Tendencias Tecnologicas - Mecatronica

Peter Hehenberger · David Bradley
Editors
Mechatronic
Futures
Challenges and Solutions for
Mechatronic Systems and their
Designers
Mechatronic Futures
Peter Hehenberger · David Bradley
Editors
Mechatronic Futures
Challenges and Solutions for Mechatronic
Systems and their Designers
13
Editors
Peter Hehenberger David Bradley
Institute of Mechatronic Design School of Science, Engineering
and Production and Technology
Johannes Kepler University Linz Abertay University
Linz Dundee
Austria UK
ISBN 978-3-319-32154-7 ISBN 978-3-319-32156-1 (eBook)

DOI 10.1007/978-3-319-32156-1
Library of Congress Control Number: 2016940131
© Springer International Publishing Switzerland 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG Switzerland
Foreword
It is an honour and a pleasure to write a foreword for the Mechatronic Futures

book cleverly devised and edited by Prof. David Bradley and Dr. Peter
Hehenberger. This contribution is written on behalf of numerous colleagues in the
UK and the wider world who have supported the Mechatronics Forum in its mis-
sion and activities. Founded in 1990, this forum was formed as an interest group
in the UK and is currently sponsored by and part of the Institution of Mechanical
Engineers (IMechE), London. Over the years since its formation, it has enabled
professional engineers from industry and academia to share innovative ideas in
this field and has acted as body to champion the discipline.
Mechatronics is a domain which spans most of my professional career and has
been at the centre of fascinating technological developments and trends in the
modern world. Mechatronic Futures reviews the origins of the discipline, explains
the technological changes that have emerged and explores how mechatronics will
further develop in future years, the challenges it will face and how it might need
to respond in helping to address some of the grand challenges that are facing
mankind.
The concept of mechatronic systems was first used in Japan in the 1960s by
Tetsuro Mori to reflect the emerging role of electronic components used in the
control and operation of what had previously been inherently mechanical sys-
tems. The Mechatronics Forum came into existence at a meeting held at IMechE’s
London headquarters on 30 October 1990, attended by over seventy engineers
excited by an interest in the emerging discipline. It was the first organisation in
the Western World that recognised the importance of mechatronics and began
to promote it. Although the word mechatronics has been around since the late
1960s, it was only in the early 1990s that it was used to any great extent in the
UK. However, during the 1990s, with the activities of the Mechatronics Forum,
the term mechatronics and the engineering discipline that it encompasses became
widely recognised.
Mechatronics today extends beyond the integration of mechanical, electronic
and computer engineering. Many engineers now see it as embracing a wider range
of engineering activities, from design through manufacture to the market place.
v
vi Foreword
Hence, they regard mechatronics as a major influence in pulling together and inte-
grating the many aspects of engineering which increased specialisation has tended
to push apart from each other during the past decades. It was in an attempt to solve
this increasingly challenging problem that the Mechatronics Forum was conceived
as a first step towards the building of bridges between the many technologies, phi-
losophies and disciplines which comprise mechatronics and the professional insti-
tutions that are committed to their own particular specialised subjects. In the UK,
engineering institutions are important in sharing technical subjects between pro-
fessionals in industry and academia. They accredit undergraduate and postgraduate
courses as suitable for covering the academic components of a chartered engi-
neer’s development, and they grant Chartered status to those whose careers show
sufficient engineering responsibility and understanding to be leaders in their field.
The Mechatronics Forum for its first ten years was supported under an inter-
institutional arrangements, with secretarial and administrative services pro-
vided alternately by the Institution of Mechanical Engineers (IMechE) and the
Institution of Engineering and Technology (lET). Following this, the Forum has
been supported by the IMechE and linked with the Mechatronics, Informatics and
Control Group (MICG).
The founding Committee of the Mechatronics Forum was charged with a broad
remit including setting up and establishing a publication of a newsletter, popularising
mechatronics, focusing on educational issues and seeking ways of bringing together
all those interested in mechatronics, and especially of promoting closer links
between industry and academia. Many of these are still the remit of the Forum today,
and significant advances in a number of areas have been facilitated through the aus-
pices of the Forum. The Mechatronics Forum Committee has included a number of
members from outside the UK, to help with the internationalisation of the Forum
and its activities as illustrated by most of the biennial international conferences being
hosted outside the UK.
The original founding members of the Forum were Prof. Jack Dinsdale, the first
chair of the organisation, Prof. Jim Hewit and Prof. David Bradley, each of whom
were made Honorary Life Presidents of the organisation.
The showcase activity of the Mechatronics Forum since its formation has been
the series of biennial international conferences, the first and longest standing con-
ference on mechatronics in the world, featuring important contributions from
around the globe. The very first conference was organised by Prof. David Bradley,
whilst working at Lancaster University. The conferences have been an excellent
means of sharing mechatronic ideas, thinking and applications more widely.
Hosts and venues for the fifteen conferences were:
1989: Lancaster University, UK
1990: Cambridge University, UK
1992: University of Dundee, Scotland, UK
1994: Technical University of Budapest, Hungary
1996: University of Minho, Portugal
1998: University of Skövde, Sweden
Foreword vii
2000: Georgia Institute of Technology, Atlanta, USA

2002: University of Twente, The Netherlands
2004: Middle East Technical University, Ankara, Turkey
2006: Penn State University, Great Valley, Pennsylvania, USA
2008: University of Limerick, Ireland
2010: ETH, Zurich, Switzerland
2012: Johannes Kepler University, Linz, Austria
2014: University of Karlstad, Sweden
2016: Loughborough University, UK
Another highlight in the calendar of the Mechatronics Forum is the Prestige
Lecture. Every year, an eminent speaker is invited to the headquarters of IMechE
to deliver a topical and sometimes challenging lecture to the members. The follow-
ing are some of the Prestige Lecturers to date:
• The Role of Xero-Mechatronics in New Product Development, Dr. John F. Elter,
Xerox Corporation, April 1995
• Advances in Mechatronics: the Finnish Perspective, Vesa Salminan, FIMET,
Finland, May 1996
• The Industrial Benefits of Mechatronics: the Dutch Experience, Prof. Job van
Amerongen, University of Twente, the Netherlands, May 1997
• Virtual Worlds—Real Applications: Industrial and Commercial Developments in
the UK, Prof. Bob Stone, University of Birmingham, May 1998
• Mechatronics Solutions for Industry, Prof. Rolf Isserman, University of
Darmstadt, April 2000
• Intelligent Mechatronics: Where to go? Prof. Toshio Fukuda of Nayaga
University, July 2001
• Bionics: New Human Engineered Therapeutic Approaches to Disorders of the
Nervous System, Prof. Richard Norman, University of Utah, July 2003
• GM’s Approach to Eliminating Complexity and Making the Business More
Successful, Dr. Jeffrey D. Tew, General Motor’s R&D Center, September 2004
• Mechatronic Design Challenges in Space Robotics, Dr. Cock Heemskerk and
Dr. Marcel Ellenbroek, Dutch Space, June 2005
• Cyborg Intelligence: Linking Human and Machine Brains, Prof. Kevin
Warwick, University of Reading, May 2006
• Iterative Learning Control—From Hilbert Space to Robotics to Healthcare
Engineering, Prof. Eric Rogers, University of Southampton, March 2007
• World Water Speed Record Challenge—The Quicksilver Project, Nigel
McKnight, Team Leader and Driver, Quicksilver (WSR) Ltd., May 2008
• Meeting the Challenges and Opportunities of Sustainability through
Mechatronics Product Development, Prof. Tim McAloone, Technical University
of Denmark, Denmark, May 2009
• I’m a Control Engineer: Ask me what I do? Prof. Ian Postlewaithe, University of
Northumbria, March 2010
viii Foreword
• Sports Technology: The Role Of Engineering In Advancing Sport, Dr. Andy

Harland, Sports Technology, Loughborough University, September 2011
• Modelling, Control and Optimisation of Hybrid Vehicles, Lino Guzzella, Swiss
Federal Institute of Engineering, October 2013
• The Past, Present and Future of Mechatronics: A Personal Perspective,
Professor Emeritus David Bradley, Abertay University, February 2015.
In addition to the biennial international conference series and the Prestige
Lectures, the Forum has supported and facilitated many other activities related to
mechatronics and its wider promotion including:
• Technical visits to industrial and academic organisations where these are open
to members.
• Technical seminars and workshops focussed on particular aspects of the
discipline.
• Mechatronics Student of the Year Award, a competition open to final-year
degree students and master’s level students based upon their final project
dissertation.
To conclude the Mechatronics Forum is part of the Mechatronics, Informatics
and Control Group of the Institution of Mechanical Engineers. It has had an
important role in popularising mechatronics in the UK and, beyond, focusing on
educational issues and promoting linkages between industry and academia, and
seeking ways of bringing together all those interested in mechatronics.
Over the past twenty-five years or so, this has been encouraged through a vari-
ety of conferences, lectures, seminars, events and visits to industry and research
centres. Mechatronics today is significant in industries worldwide. In our global
environment, collaboration across the world is valuable in moving towards sus-
tainable systems and solutions to societal challenges. The biennial conferences in
particular have proven to be very fruitful in sharing ideas and stimulating debate
between those at the forefront of mechatronics development which we hope will
benefit societies across the world. Mechatronic future sets out a vision and narra-
tive of how some of the future societal challenges might be addressed and the role
mechatronics has to contribute to this.
Prof. Philip Moore

Falmouth University
Contents
1 Mechatronic Futures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
David Bradley and Peter Hehenberger
2 Mechatronics Disrupted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Maarten Steinbuch
3 Future Challenges in Mechatronics. . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Nicolas Albarello, Alexandre Arnold and Marc Budinger
4 TiV-Model—An Attempt at Breaching the Industry Adoption
Barrier for New Complex System Design Methodologies. . . . . . . . . . 41
Craig Melville, Xiu-Tian Yan and Lixiang Gu
5 Digital Twin—The Simulation Aspect. . . . . . . . . . . . . . . . . . . . . . . . . . 59
Stefan Boschert and Roland Rosen
6 Design Processes of Mechatronic Systems . . . . . . . . . . . . . . . . . . . . . . 75
Matthieu Bricogne, Julien Le Duigou and Benoît Eynard
7 Design of Biomimetic Soft Underwater Robots. . . . . . . . . . . . . . . . . . 91
Aiguo Ming and Wenjing Zhao
8 Improving the Robustness of Mechatronic Systems . . . . . . . . . . . . . . 113
Antonio Lanzotti and Stanislao Patalano
9 Integrated Manufacturing: The Future of Fabricating
Mechatronic Devices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Nicholas Fry, Rob Richardson and Jordan H. Boyle
10 From Mechatronic Systems to Cyber-Physical Systems:
Demands for a New Design Methodology?. . . . . . . . . . . . . . . . . . . . . . 147
Peter Hehenberger, Thomas J. Howard and Jonas Torry-Smith
11 The Internet of Things: Promise of a Better Connected World . . . . . 165
George R.S. Weir
ix
x Contents
12 Home Technologies, Smart Systems and eHealth. . . . . . . . . . . . . . . . . 179

Jorge Azorin-Lopez, Andres Fuster-Guillo, Marcelo Saval-Calvo
and David Bradley
13 The Changing Landscape of Enterprise ICT—Responding
to New Possibilities and User Demands . . . . . . . . . . . . . . . . . . . . . . . . 201
Christopher Milne and Steve Watt
14 Engineering Design for Mechatronics—A Pedagogical
Perspective. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Simeon Keates
15 Mechatronics Education: Meeting Future Need . . . . . . . . . . . . . . . . . 239
David Russell
16 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Peter Hehenberger and David Bradley
Editors and Contributors
About the Editors
Peter Hehenberger is an assistant professor and deputy head at the Institute of

Mechatronic Design and Production at the Johannes Kepler University of Linz
(JKU) in Austria. He received his diploma degree and his doctorate degree in Me-
chatronics from JKU in 2000 and 2004, respectively.
His core competencies cover model-based mechatronic system design and
include mechatronic design, systems engineering, design processes and model-based
design methods. He has published over 70 peer-reviewed papers in international
journals and conference proceedings and been guest editor for two journals’ special
issues. In addition, he has authored and co-authored 3 books related to the field.
He is presently a member of the Scientific Network of Linz Center of
Mechatronics (LCM), Austria, and leads several international research projects related
to the research area “Process Modelling and Mechatronic Design” at the LCM.
Peter Hehenberger is also one of the initiators of SmaPro,1 a qualification network
for the future challenges of Industry 4.0 together with 13 partners—four academic
partners, four small and mid-sized companies as well as five large companies.
He lectures on the topics in product development, model-based systems engi-
neering, mechanical engineering, mechatronic design, computer-aided design,
simulation and manufacturing. He is currently co-supervising 6 doctoral students
and 5 master theses by research students.
Among other scientific service activities, Peter Hehenberger organised the
International Workshop Series on Mechatronic Design (Linz 2012, Paris 2013,
Leuven 2014). He is also member of VDI, ASME, IFIP WG5.1 (“Global Product
development for the whole life-cycle”) and Design Society, where he co-chairs a
Special Interest Group (SIG) on Methodologies for Design, Integration, Modelling
and Simulation of Cyber Physical Systems.
1
Smart Production: Machine data analysis and interpretation in production.
xi
xii Editors and Contributors
Between February and April 2015, he was an invited visiting professor at the
Universite de Technologie de Compiegne, Departement Genie des Systemes
Mecaniques, France.
David Bradley is currently professor emeritus at Abertay University. He began his
academic career when he joined Lancaster University in 1972 and has been engaged
with and involved in aspects of mechatronics since the mid-1980s and was responsi-
ble, with Prof. Jim Hewitt and Prof. Jack Dinsdale, for the establishment of the UK
Mechatronics Forum, now the Mechatronics Forum.
In the course of a period of some 30 years at Lancaster University, the
University of Wales Bangor and Abertay University, he has been involved in vari-
ous aspects of the design, manufacture and operation of mechatronic systems for
applications ranging from intelligent robots to medical and telecare systems and
design methods. This included the establishment at Lancaster University in the late
1980s of some of the first mechatronics degree programmes, at both undergradu-
ate and master’s levels, in the UK as well the EPSRC-funded Engineering Design
Centre which had mechatronics as a focus.
Most recently, he has been interested in exploring the interaction between
mechatronics, the Internet of Things and Cyber-Physical Systems, and in par-
ticular how the design process can be adapted at both the systems and component
levels to accommodate context-dependent systems elements derived from cloud-
based technologies. This includes the means by which the design process enables
the development of participatory systems structured around user need, essentially
placing the user, who may be interested only in context, at the heart of the process
of system integration, and including issues related to user privacy and security.
He is also working in an advisory or consultative capacity with research groups
in the UK and elsewhere on projects to develop advanced prostheses and evaluate
mechatronics design methods.
A fellow of the Institution of Engineering and Technology (IET), David is the
author, co-author or contributing editor for 7 textbooks, including 3 on mechatron-
ics, some 50 journal papers and book chapters along with over 120 conference
papers. He has also been responsible for the development of, and teaching on,
mechatronics courses in the UK since the mid-1980s and has lectured on and acted
as consultant for similar courses in, among others, the USA, Singapore, South
Africa, Indonesia and Colombia.
Contributors
Nicolas Albarello is a researcher in the Systems Engineering Department of Airbus

Group Innovations. He received his Ph.D. in Sciences for the Engineer from Ecole
Centrale Paris in France. Since 2012, he has been contributing to several research
projects in the field of systems engineering but also to operational projects for civil
aircrafts and space systems. His main research topics include optimisation of systems
(architecture, sizing, control), modelling and simulation, and decision aid in design.
Editors and Contributors xiii
Alexandre Arnold is a research engineer at Airbus Group Innovations in the

Model-based System and Software Engineering team. He graduated in 2010 with
a double diploma in engineering from the Ecole Centrale Lyon in France and the
Technical University Munich in Germany, specialising in aeronautics and aero-
space. He has since been working on various research projects, as well as opera-
tional projects for civil/military aircrafts and space division. The research topics
have mainly been focused on MBSE, modelling and simulation, formal verification
and systems of systems.
Jorge Azorín-López received a degree in Computer Engineering in 2001 and a
Ph.D. degree in Computer Science at the University of Alicante (Spain) in 2007.
Since 2001, he has been a faculty member of the Department of Computer Tech-
nology at the same university, where he is currently an associate professor and the
deputy director of Research.
He was awarded the Post-doctoral Research Fellowship “Automatic visual
inspection of shape defects on specular surfaces. Methods to compensate low sen-
sitivity of 3D image acquisition and reconstruction techniques” by the Spanish
Ministry of Science and Education for research at University of Edinburgh.
He has worked in 16 research projects and has published more than 50 papers
on computer vision and computer architecture in several journals, conferences and
book chapters. He has also served as a reviewer to numerous scientific journals
and international conferences. Currently, he is associate editor of the IJCVIP jour-
nal. Current research interests include human behaviour analysis using computer
vision, visual inspection and 3D modelling and analysis using computer vision.
Stefan Boschert graduated in physics from the University of Freiburg. After gain-
ing his Ph.D. in 2000, he worked for an engineering company in Basel, Switzerland,
on mechanical and hydro-dynamical modelling and simulation. Since 2004, he has
worked at Siemens Corporate Technology in Munich, Germany, in the technology
field of automation and control.
As a Senior Key Expert Engineer he is responsible for simulation-based system
development, which includes multiphysics simulation, process models and inte-
grated data and simulation model-based approaches. He has expertise in the set-up
and implementation of mechatronic design processes and their extension to cyber-
physical systems. Further, he was the manager of the department’s contribution to
several national and international publicly funded projects.
Jordan H. Boyle is a lecturer in the School of Mechanical Engineering, at the
University of Leeds, and is also a founding member of the EPSRC National Facility
for Innovative Robotic Systems, an advanced manufacturing facility. He obtained
his B.Sc. and M.Sc. degrees in Electrical Engineering from the University of Cape
Town, South Africa, before moving to the University of Leeds for a Ph.D. in Com-
putational Neuroscience in the School of Computing.
Jordan’s research lies on the interface of engineering and biology, and his main
interest is in using biological inspiration to develop locomotion and navigation strat-
egies for mobile intelligent robots. Since the phenomenal complexity of biological
xiv Editors and Contributors
systems makes it extremely challenging to decipher the mechanisms involved—a

crucial first step in the biologically inspired design process—Jordan prefers to focus
on relatively simple invertebrate systems. To date, his specific research projects
have included developing a miniature mobile robot for use in colonoscopy and a
snake-like robot controlled by a model of the C. elegans nervous system.
Robotics applications underlie Jordan’s interest in advanced manufacturing.
Indeed, one of the major barriers to the wider uptake of mobile autonomous robots
is cost. If robots could be made cheaply and in large volumes, they would be
deployed more widely, more frequently and in greater numbers.
Matthieu Bricogne is an assistant professor at the Mechanical Systems Engineer-
ing Department/Roberval Laboratory (UMR 7337) in Université de Technologie
de Compiègne (UTC). He obtained his M.Sc. in Mechanical Engineering and a
Mechanical Engineering Degree at UTC in 2004.
He has worked for 5 years for Dassault Systèmes, an industrially leading
company specialising in 3D and Product Lifecycle Management (PLM) soft-
ware. He obtained his Ph.D. degree in Mechanical Engineering from the Ecole de
Technologie Supérieure in Montréal (ETS) and UTC in 2015. His main research
topics are collaborative design, mechatronics system design and knowledge-based
engineering within a PLM environment.
Since 2013, he has been vice-director of the ANR LabCom DIMEXP (DIgital
MockUp multi-EXPertises) joint laboratory UTC/DeltaCAD, whose main objec-
tive is to develop methods and models to improve the integration of multidiscipli-
nary expertises and heterogeneous data in advanced digital environments.
Marc Budinger is an alumni of the ENS Cachan (Paris) in Applied Physics, has
a Ph.D. in Electrical Engineering (2003) and a HdR (habilitation) in mechanical
engineering (2014) from Toulouse University.
Marc is an associate professor in the Mechanical Engineering Department of
INSA Toulouse. His current research topic in the ICA laboratory concerns the
model-based design of electromechanical and piezoelectric actuators and actua-
tion systems. He is currently investigating the conjoint use of estimation/predic-
tion models, 0D/1D simulation models, reliability models and metamodels from
3D FEM simulation for design and optimisation of mechatronics systems.
Benoît Eynard is currently the director of the Institute for Mechatronics at the
Université de Technologie de Compiègne—UTC, France. In 2007 he joined UTC as
a full professor, leading the Department of Mechanical Systems Engineering until
2012. He is also member of the UMR CNRS/UTC 7337 Roberval. Previously, he
was an assistant professor at the Université de Technologie de Troyes where he man-
aged the M.Sc. programme in Information Technology for Mechanical Engineering.
Since 2013, he has been general chairman of the academic group of French
Society of Mechanical Engineering (AFM) dealing with factories of the future: inte-
grated design and advanced manufacturing also known as AIP-PRIMECA network.
He is also a member of the IFIP Working Group 5.1 “Global Product develop-
ment for the whole life-cycle” and of the Design Society where he currently leads
the Special Interest Group on “Design Methods for Cyber-Physical Systems”.
Editors and Contributors xv
In 1999, he obtained a Ph.D. in Computer-Integrated Manufacturing from the

University of Bordeaux. Now, he is a recognised researcher in product lifecycle
management, collaborative design, systems engineering, mechatronic design,
digital factory, manufacturing process management, eco-design and sustainable
manufacturing. He has published over 50 papers in international journals and 150
international conferences. He also has been guest editor for 10 journal special
issues and academic books.
Nicholas Fry is undertaking a Ph.D. research degree within the Leeds EPSRC
National Facility for Innovative Robotic Systems, UK. He received his B.Eng. and
M.Eng. degrees in Mechatronics and Robotics from the University of Leeds. Nicho-
las is researching advanced manufacturing techniques and has recently developed
a novel 3D printing process, using industrial robotic arms, with the aim of autono-
mously embedding mechatronic components within rapidly fabricated structures.
Andres Fuster-Guillo received the B.S. degree in Computer Science Engineering
from Polytechnic University of Valencia (Spain) in 1995 and the Ph.D. degree in
Computer Science at the University of Alicante (Spain) in 2003.
Since 1997, he has been a member of the faculty of the Department of
Computer Technology at the University of Alicante, where he is currently a pro-
fessor. He has worked in more than 15 research projects and has published more
than 40 papers on computer vision and computer architecture.
His current areas of research interest include vision systems to perceive under
adverse conditions, 3D vision systems, automated visual inspection and artificial
neural networks.
He was deputy coordinator of the Polytechnic School at the University of
Alicante for seven years. Currently, he is director of the Secretariat for Information
Technology at the University of Alicante, where he has coordinated and partici-
pated in several strategic technology projects: Open University (transparency por-
tal and open data), UACloud, and Smart University, among others.
Thomas J. Howard is an associate professor at the Technical University of Denmark
(DTU) after gaining his Ph.D. in Engineering Design from the University of Bath,
UK. His main field of expertise is in robust design and more broadly engineering de-
sign research. He currently heads the robust design group at DTU and is the current
chairman of the Design Society’s Special Interest Group on Robust Design.
Tom has numerous past and present Ph.D. students and has attracted large
amounts of research funding from both public and private sources including
investment for the Novo Nordisk-DTU Robust Design Programme of around
€1.5M. He sits as an elected member of the Design Society’s advisory board and
has received several accolades such as the DTU Innovation Prize 2014 and the
BMW European Scientific Award. He has also published work in many top engi-
neering design journals.
In addition to the research on robust design, Tom has a keen interest in inno-
vation and technical entrepreneurship. He has been a keynote speaker at the
European Innovation Academy several times and has been responsible and instru-
mental in numerous spinout companies.
xvi Editors and Contributors
Simeon Keates is professor and deputy pro-vice-chancellor of the Faculty of En-

gineering and Science at the University of Greenwich. He was formerly chair of
HCI and head of School of Engineering, Computing and Applied Mathematics at
Abertay University in Dundee and associate professor at the IT University of Co-
penhagen, where he lectured in the design and digital communication study line. He
obtained his Ph.D. from the University of Cambridge, where he also worked as an
Industrial Research Fellow in the Engineering Design Centre.
After leaving Cambridge, he moved to the USA and joined the Accessibility
Research Group at the IBM TJ Watson Research Center before moving to Boston
and working at ITA Software (now part of Google) as a usability lead design-
ing interfaces for Air Canada. Simeon also has an extensive history of consul-
tancy, with clients including the Post Office (Royal Mail), the US Social Security
Administration, the UK Department of Trade and Industry, Danish Broadcasting
Corporation (DR) and Lockheed Martin.
Antonio Lanzotti is the founding director of the Fraunhofer IDEAS (Interactive
DEsign And SimulaTION) LABORATORY (www.ideas.unina.it) and is leading a
research group on design methods at the University of Naples Federico II. He re-
ceived his master’s degree in Mechanical Engineering (summa cum laude) from
Università degli Studi di Napoli Federico II in 1985.
From 1986 to 1987 he worked for the IT Department of Aeritalia. In 1990 he
received a Ph.D. in Material and Production Technology from the University of
Naples Federico II.
From 1992 to 1997, he undertook teaching and research activities at the
Department of Aerospace Engineering as researcher and then from 1997 in a
teaching position as associate professor. In 2001 he was appointed Full Professor
in Design Methods of Industrial Engineering at University of Naples Federico II.
He now teaches “Product Design and Development” and “Engineering drawing”.
He has been the only teacher of “Statistics for Innovation” not belonging to the
“Experimental Statistics for Engineering and Technology” Scientific Sector.
He is associate editor in chief of the International Journal on Interactive
Design and Manufacturing published by Springer. He is a reviewer for interna-
tional journals, and he has been invited to lecture in international conferences
and has promoted and organised many international conferences as a mem-
ber of Steering and Scientific Committees. He was the President of the Italian
Association of Product Design from 2004 to 2011. Further, he is the past presi-
dent of the Italian Scientific Sector on Design Methods for Industrial Engineering
(SSD ING-IND/15). Finally, he is responsible for the international agreement
between the University of Naples Federico II and TU Chemnitz, ESTIA of
Biarritz and University of Bartin, Turkey.
Julien Le Duigou is an associate professor at the Department of Mechanical
Systems Engineering of Université de Technologie de Compiègne and researcher
at Roberval Mechanical Laboratory (UMR CNRS 7337) and Labex MS2T “Control
of Technological Systems-of-Systems” research centres. He obtained his mechanical
engineering degree and his M.Sc. in Mechatronics in 2006 from Supmeca engineering
Editors and Contributors xvii
school and University of Toulon, respectively, and a Ph.D. in Mechanical Engineer-

ing in 2010 from Ecole Centrale engineering school.
He is also member of the IFIP working group 5.1 dealing with “Global
Product development for the whole life-cycle” and of the IEEE Systems Man
and Cybernetics Society. Currently, his research interests include product lifecy-
cle management, knowledge management, interoperability and product/process
integration. He published more than 50 papers in refereed international journals,
books and conferences.
Craig Melville is a Ph.D. student studying in the Space Mechatronic Systems Tech-
nology Laboratory (SMeSTech) at the University of Strathclyde in Glasgow, having
received his M.Eng. in Product Design Engineering in 2013 from the university’s
Department of Design, Manufacturing and Engineering Management (DMEM). His
terrestrial design projects have involved a range of intra-discipline skills such as
computer-aided design and simulation, structural analysis and design optimisation.
The projects he has been involved in allowed him to work with both Scottish
SMEs and large international organisations; one project for Jaguar Land Rover
allowed them to make design changes capable of saving up to £1,000,000 in
assembly and manufacturing costs over 3 years.
Craig is currently focused on work relating to spacecraft and space robotic
design for his Ph.D. studies, which he is due to finish in 2017. His work in the area
focuses on the development of a design methodology for flexible use within space
and complex design organisations. Other interesting realms of work include the
design and optimisation of a multifunctional spacecraft interface, the development
of universal interface design principles and the role of virtual reality and video
games in conceptual/low TRL design phases.
Christopher Milne is associate chief information officer (Information Assurance
and Governance) at the University of St. Andrews.
He gained a B.A. (Hons) in Librarianship and Information Studies at the Robert
Gordon University (RGU) in 1994 before obtaining a PgDip in Information
Analysis (1995). Following a spell as an IT trainer, Chris returned to RGU, where
he was responsible for developing IT skills teaching in undergraduate and post-
graduate modules.
In 1998, Chris joined the then University of Abertay Dundee, where he had
a rich and varied career, predominately in the field of information management
(academic librarian) and latterly as the officer responsible for corporate informa-
tion and risk management across the university, as a Deputy Head of Information
Services. Whilst at Abertay he read for a Masters in Records Management at
Northumbria University (2007). He has published in the fields of information
literacy—integrating core skills into the curriculum (2004), and records man-
agement—advocating the use of records management information retrieval phi-
losophies as a means of improving access to information from corporate Websites
(2007) and (2010).
In his current role at the University of St. Andrews, he is responsible for infor-
mation governance and records management across the institution. He has a
xviii Editors and Contributors
business focused view of legislation and regulation—in that it should aid institu-
tions in their mission and development and not hinder them via obsessive com-
pliance and the introduction of unnecessary and/or artificial barriers. As part of
the senior management team of the University’s IT Services, he is also responsible
for providing leadership and direction as IT Services emerges from a period of
reorganisation. From 2015, he also gained responsibility for the day-to-day man-
agement of the university’s complaint management function, developing organisa-
tional learning by assessing complaint contributory factors.
Aiguo Ming received the M.S. degree in Precision Engineering from Yamanashi
University, Japan, in 1987, and the Ph.D. degree in Precision Machinery Engineer-
ing from the University of Tokyo, Japan, in 1990.
He is currently a professor in the Department of Mechanical Engineering and
Intelligent Systems, The University of Electro-Communications, Tokyo, Japan.
His current research interests include biomimetic hyper-dynamic robotics, bio-
mimetic soft robotics, high-speed robotic hand systems with high-speed sensors,
application of mobile manipulator systems and precise measuring mechatronic
systems.
Philip Moore became pro-vice-chancellor (R&I) at Falmouth University in 2012.
He leads all research and development, and is also chair of the Centre for Smart
Design. Philip has been a visiting professor at University of Skövde, Sweden, since
1994 and on the Board of Directors of the Smart Home and Building Association
(SH&BA). Phil was previously director of Research at De Montfort University
(DMU) and director of the Mechatronics Research Centre and architect of the Intel-
ligent Machines and Automation Systems (IMAS) laboratory. Phil studied Produc-
tion Engineering and Management at Loughborough University whilst working in
the automotive industry. He then studied for a Ph.D. in Robotics, whilst working in
research at Loughborough, before joining as an academic member of staff, where he
was a founder member of the Manufacturing Systems Integration (MSI) Research
Institute.
Phil’s own research focuses on smart digital technologies and their integration
with sustainable design to address some of the societal grand challenges such as
climate change, energy resilience, health and well-being and the ageing society.
Other research interests include digital manufacturing and automation.
Phil was responsible for the founding of Digital Games at Falmouth and the
associated research in computer games technology and was the architect of the
successful European Research Area (ERA) Chair bid for €2.4 million to the FP7
programme in Digital Games research. Phil has supervised some 40 Ph.D. pro-
grammes to completion and has been external examiner on nearly 50 occasions.
Philip has won research funding from UK and EU agencies—EPSRC, ESRC,
InnovateUK, EU FP, British Council, The Royal Society, RDA’s and Industry.
He has established a network of research collaborators in industry and academia in
UK, Sweden, Germany and Italy.
Phil is vice-chair of Mechatronics, Informatics and Control Group, IMECHE,
and chair of the Mechatronics Forum.
Editors and Contributors xix
Stanislao Patalano is an associate professor in Design and Methods of Engineering

Design at Department of Industrial Engineering, University of Naples Federico II.
He gained a master’s degree in Mechanical Engineering in 1997, Ph.D. title in
“Machine Design and Construction” in 2001, at the University of Naples Federico II.
From 1999 until 2010, he was researcher in the Scientific Area of Design and
Methods of Engineering Design. In 2006, he was visiting professor at the Institut
Supérieur de Mécanique de Paris—SUPMECA, France, working on tolerance design.
He has published more than 90 papers in International Conference Proceedings,
International Journals and National Journals. His research interests include CAD
methodologies and knowledge-based engineering, variational analysis, tolerance
design and design for assembly, and mechatronic systems design.
He has been member and coordinator of research groups within research pro-
jects funded by national public agencies (MIUR, Campania Region). From 2011 to
2014, he has been scientific responsible for the research activities of the Department
of Industrial Engineering within “PON Project 01_01268: Digital Pattern
Development: a pattern driven approach for industrial product design, simulation
and manufacturing”, in cooperation with FIAT Item, Fiat Chrysler Automobiles
(FCA), Ansaldo-Breda, Ansaldo STS, SMEs (Technodesign, Tecnosistem, STEP
Sud Mare).
He has scientific responsible for the COGITO (COmputer GeometrIc Modelling
and SimulaTiOn) Laboratory at the Department of Industrial Engineering of
the University of Naples Federico II. COGITO Laboratory is a part of IDEAS
(Interactive DEsign And Simulation) Laboratory, established between University
Federico II and Fraunhofer IWU.
He is currently professor of Geometric Modelling and Virtual Prototyping and
Engineering Drawing at the University of Naples Federico II.
From 2012 to present, Promoter and Responsible for the University of Naples
Federico II of the Academic Course aimed to a double master’s degree: “Master
Degree in Mechanical Engineering for Designing and Manufacturing” and
“Diplôme d’Ingénieur Supméca”.
Rob Richardson is professor of Robotics at the School of Mechanical Engineering,
University of Leeds, UK, and a fellow of the IMechE. He is director of the Leeds
EPSRC National Facility for Innovative Robotic Systems, a world class, facility for
designing and creating robotic systems initiated in August 2013. He is also director
of the Institute of Design, Robotics and Optimisation at the School of Mechanical
Engineering. He is co-director of the UK EPSRC cities grand challenge, awarded
in November 2015 with the aim of using robotics to perform minimally invasive
infrastructure repair.
He held a prestigious research contract to explore the Great Pyramid of Giza,
Egypt, using robotic technology and has discovered writing in the Great Pyramid
that was hidden for thousands of years.
His current research interests focus around the fabrication of complex robotic
systems with application to exploration robotics for safety and security including:
inspection and repair of city infrastructure; protecting life by accessing confined
xx Editors and Contributors
spaces to find survivors of natural disasters; gathering information on terrorist and

military threats; inspecting inaccessible industrial plants; or processes utilising
enhanced sensing and networking.
Roland Rosen studied mathematics with a minor in economics at the Technical
University of Munich. After joining Siemens in 1989, he worked in different posi-
tions in the area of modelling, simulation and optimisation of technical systems,
plants and infrastructure solutions. Currently, he is at Siemens Corporate Technol-
ogy as a Principal Key Expert Research Scientist responsible for the international
focus area of modelling and simulation.
He is involved in different activities related to “Industrie 4.0”, and he is a mem-
ber of expert committees of VDI/VDE Society for Measurement and Automatic
Control (GMA). Roland Rosen works on the realisation of simulation-based meth-
ods for mechatronic systems and complex plants. His specific interest is in the
seamless integration of simulation into all lifecycle phases of system development
and the use and reuse of simulation models and technologies during operation and
services.
David Russell was awarded his Ph.D. in recognition of his research at the UK
Atomic Energy, the University of Manchester and the then Liverpool Polytechnic.
His thesis included the design of an intelligent controller for a novel fast reactor us-
ing a fluidic control schema.
Dr. Russell is a professor of engineering at Penn State Great Valley where
has spent the last 30 years. Dr. Russell was past chair of the campus IT Advisory
Board and was a founder member of the University Committee for Enterprise
Integration, which is part of the SAP University Alliance programme. Whilst sen-
ior division head for fourteen years, Professor Russell designed and implemented
two professional postgraduate degrees in Software Engineering and Systems
Engineering which are now available on-campus and online.
He has taught at the undergraduate and post graduate level in the UK, USA
and Asia over the past half-century and has created, managed and taught systems
engineering, information technology and computer engineering activities includ-
ing scholarly and professional international conferences. He has delivered invited
and plenary lectures in over thirty countries. Dr. Russell serves the editor for the
Americas for the International Journal of Advanced Manufacturing Technology, a
Springer publication.
Dr. Russell is a past fellow of the Institute of Electrical Engineering and fellow
of the BCS (British Computer Society) and a current fellow of the Institute of
Mechanical Engineers (IMechE) and a Chartered Engineer.
Prior to Penn State Dr. Russell held several positions as a technical executive
with factory systems integration companies in the USA and the UK and has held
several faculty appointments in other universities. He is the author over 120 tech-
nical papers and an active international member of the Mechatronics Forum, an
IMechE scholarly society.
Editors and Contributors xxi
Marcelo Saval-Calvo received the degree in Computer Engineering in 2010 and

the M.Sc. in 2011 from the University of Alicante (Spain) with a master’s thesis in
the area of computer vision. He received his Ph.D. in 2015 from the University of
Alicante focused on 3D registration of rigid and deformable subjects.
He is currently working in the Computer Technology Department at the
same university. He was awarded a research fellowship by Regional Ministry of
Education, Culture and Sports of Valencian Government for researching at the
University of Edinburgh. His current research interests include human behaviour
analysis using computer vision, 3D registration of rigid and deformable data.
Maarten Steinbuch is distinguished university professor at Eindhoven University
of Technology. He received the M.Sc. degree and Ph.D. degree from Delft Univer-
sity of Technology, in 1984 and 1989, respectively. From 1987 to 1998, he was with
Philips Research Laboratories in Eindhoven as a Member of the Scientific Staff.
From 1998 to 1999 he was manager of the Dynamics and Control Group at Philips
Center for Manufacturing Technology.
Since 1999, he has been full professor and chair of the Control Systems
Technology Group of the Mechanical Engineering Department of Eindhoven
University of Technology.
He has been editor in chief of IFAC Mechatronics (2009–2015), scientific direc-
tor of the Research Centre High Tech Systems of the Federation of Dutch Technical
Universities, director of the TU/e Graduate Program Automotive Systems, initiator
of the TU/e Bachelor Automotive, member of the board of AutomotiveNL, member
of the Formule E team, member of the advisory board of the High Tech Institute
and of Nobleo Technology BV, co-founder of MI-Partners BV, co-founder of
Mechatronics Academy BV, co-founder of Preceyes BV and founder of Steinbuch
in Motion BV. He is CEO of Medical Robotics Technology BV.
Since 2014, he is also the scientific director of the new TU/e High Tech Systems
Center HTSC. In 2003, 2005, 2008 and 2015, he obtained the "Best-Teacher" award
of the Department of Mechanical Engineering, TU/e. In 2008 as well as in 2014,
his research group obtained the QANU excellence rating [5555]. In 2013 he was
appointed distinguished university professor at TU/e. In 2015 Maarten received the
KIVI Academic Society Award. His research interests are modelling, design and
control of motion systems, robotics, automotive powertrains and of fusion plasmas.
Jonas Mørkeberg Torry-Smith has been involved in design of mechatronic sys-
tems for more than a decade performing research within the field as well as carrying
out mechatronic product development in industry.
He holds a Ph.D. in systematic design of mechatronic products. The research
has been aimed at procedures for mechatronic development integrating the
mechanical, the electronics and the software engineering discipline. The research
also comprises proposals for methods for modelling mechatronic concepts across
the disciplines elucidating central conceptual dependencies.
From 2002 to 2014 Jonas Torry-Smith was affiliated to IPU—a consulting
company—working with mechanical and mechatronic development as a con-
sultant for Danish and international companies. The range of projects comprises
xxii Editors and Contributors
innovation projects, concept scouting as well as total turn-key projects. The

projects have been carried out as project manager, team member or as specialist.
In addition, on-the-job training has been performed for international companies
based in Denmark. He has recently switched to the Pharma industry working for a
large international med-tech company.
Steve Watt is chief information officer at the University of St. Andrews. A native
of Scotland he was born in Dundee in 1971 and currently lives north of Dundee in
the county of Angus.
Steve was educated at the then University of Abertay Dundee where he gained
a Bachelor of Science degree in Building Engineering and Management, a Master
of Science degree in Information Technology and later a Master of Business
Administration degree.
Steve is a European engineer, chartered engineer, chartered IT professional and
a fellow of the British Computer Society and has 20 years’ experience working
within the information technology field.
He has worked primarily in the Higher Education sector in a variety of roles
which have included Network Analyst, IT Manager, Deputy Head of Information
Services and Chief Information Officer (CIO).
Steve has also worked in the private sector for Low & Bonar plc in both Europe
and North America. In this role, he was responsible for the coordination and deliv-
ery of corporate IT provision at around 52 locations which included the provision-
ing and management of wide area networking, telecommunications, groupware
and systems management working with outsourced partners centralising the core
ICT support for all sites to the UK. He has also worked as a consultant across a
range of industries.
In his current role as CIO at the University of St. Andrews, he is responsible for
the IT services and change management functions comprising of around 90 staff
and is a member of the University Senior Management Team. He is responsible for
the strategic development of all IT provision, information assurance and govern-
ance and business transformation activities.
He is a member of the Board of Management of Dundee and Angus College,
chair of the Higher Education IT Directors in Scotland Group and a member of the
Scottish Government Technical and Design Board. Steve is a frequent contributor
at IT conferences including as conference chair and panellist.
George R. S. Weir is a lecturer in Computer and Information Sciences at the Uni-
versity of Strathclyde in Glasgow, Scotland. He holds academic qualifications from
three Scottish universities, including a Ph.D. from the University of Edinburgh. In
addition to teaching and research supervision, he has published extensively in the
areas of cybercrime, security, readability and corpus linguistics.
He is a member of the BCS Forensics and Cybercrime Specialist Group, the
Association of Computing Machinery, the British Computer Society and the
Higher Education Academy. He is also a senior member of the IEEE and a fellow
of the Winston Churchill Memorial Trust and a fellow of the Royal Society of Arts.
Editors and Contributors xxiii
Xiu-Tian Yan is professor in Mechatronic Systems Technology, the director of the

Space Mechatronic Systems Technology Laboratory (SMeSTech) and deputy director
of Strathclyde Space Institute, based in the Department of Design, Manufacture and
Engineering Management (DMEM) of the University of Strathclyde. He is the vice-
chairman of the Mechatronics Forum in the UK and chairman of the Robotics and
Mechatronics Technical Professional Network of the Institution of Engineering and
Technology.
Through over 30 years’ work and investigation, he has developed a portfolio
of research projects in mechatronic system design research including the devel-
opment and application of “mechatronic design process models”, modelling and
simulation of these systems, and techniques in prototyping these systems for vali-
dation. He has recently built a strong research team after establishing SMeSTech
research laboratory in 2012. He has managed over 40 projects in his research
fields funded by the UK government funding bodies (EPSRC, TSB/InnovateUK,
STFC), European Commission, international funding bodies and industrial
partners.
He received his Ph.D. from Loughborough University of Technology and is a
chartered engineer and a fellow of Institution of Engineering and Technology (for-
merly IEE) and a fellow of IMechE. His research interests include mechatronic
systems design, robotics especially in recent years in space robotics, Micro-
mechatronic research and multi-perspective mechatronic system modelling,
design and simulation, computer support mechatronic systems design using AI
techniques, knowledge-intensive product modelling and simulation, the proactive
computer support of product life-cycle synthesis and Product generalisation and
configuration design and constraint based insightful engineering design support.
He has organised and chaired four international conferences and published over
180 technical papers in major international journals and conferences and edited or
written 6 books in related fields. He is an invited professor or visiting professor
at 4 international universities, and he has been invited to give lectures at French,
Japanese and Chinese Universities and various research institutions. He is the
recipient of Judges Commendation Best KTP Partnership (Scotland) 2010 award
and several best paper awards. He has also been a key team member for several
industry best awards for some mechatronic products and systems.
Wenjing Zhao received the B.S. degree and M.S. degree (Mechatronics Engi-
neering) from the University of Electronic Science and Technology of China in
2007 and 2010, respectively. She received her Ph.D. degree at the Department of
Mechanical Engineering and Intelligent Systems from The University of Electro-
Communications in Japan at 2014.
During the Ph.D., her research was primarily focused on the development
of biomimetic soft underwater robots. She now is working at the College of
Mechanical Engineering in the Zhejiang University of Technology in China. Her
current research interests include soft robots, mechatronics, fluid–structure interac-
tion system, friction and sealing, and other robotic technologies.
Chapter 1
Mechatronic Futures
David Bradley and Peter Hehenberger
1.1 The Challenge
The period of over 40 years since the concept of a mechatronic system was intro-
duced by Tetsuro Mori [1] to express the growing impact that the availability of
electronic components was having on the control and operation of inherently
mechanical systems has been, and continues to be, a period of significant and
rapid technological change. In particular, there has been a shift in emphasis within
systems from hardware to firmware and software, leading to the introduction of
a wide range of consumer products structured around the use of smart devices,
many of which remain essentially mechatronic in nature in that they bring together
a core of mechanical engineering with increasingly sophisticated electronics and
software. When combined with enhanced local and remote communications, this
has led to the evolution of systems based around the ability of smart objects to
communicate with each other, and hence to effectively self-configure according to
context.
This in turn has led to the development of concepts such as Cyber-Physical
Systems, the Internet of Things and Big Data [2–11] in which interaction is driven
through the combination of smart objects and information. Referring to Figs. 1.1,
1.2 and Table 1.1, users access cloud-based structures through smart objects to
draw on resources provided by a range of, often unknown or invisible, sources.
The growth of provision represented by Table 1.1 has also led to a growth
in availability of sophisticated user systems where, for instance, smartphones
increasingly incorporate high-quality still and video imaging capability to the point
where they are now responsible for more images than conventional cameras. It has
D. Bradley (*)
Abertay University, Dundee, UK
e-mail: [email protected]
P. Hehenberger
Johannes Kepler University Linz, Linz, Austria
© Springer International Publishing Switzerland 2016 1

P. Hehenberger and D. Bradley (eds.), Mechatronic Futures,
DOI 10.1007/978-3-319-32156-1_1
2 D. Bradley and P. Hehenberger
Fig. 1.1 Cloud-based structures for the Internet of Things
Fig. 1.2 Cyber-physical systems

1 Mechatronic Futures 3
Table 1.1 Cloud functions

Applications (software as a Service—SaaS) Apps, Games, Mail, Virtual Desktop, Customer
Management, Communications, Access,
On-Demand Systems, …
Platform (platform as a Service—PaaS) Runtime Operation and Management, Databases,
Web Server, Tools, Computation, …
Infrastructure (infrastructure as a Virtual Machines, Servers, Storage, Load
Service—IaaS) Balancing, Networking, Communications, …
Fig. 1.3 Daily profile of use for mobile phones, PCs and tablets (after [12])
also led to the introduction of a range of user devices for behavioural monitoring,
smart watches and tablet computers, all of which are capable of interacting with
other smart devices through the medium of the Internet. Figures 1.3 and 1.4
together illustrate the daily profile of use for such devices [12–15]. All of the above
have implications for the design, development and implementation of mechatronic
systems, and for the future of mechatronics itself [16, 17].
In 2014, in association with the Mechatronics Forum Conference held in
Karlstad in Sweden, a number of practitioners from around the world were asked
to provide, in a single phrase, their view of the most significant challenges faced
by mechatronics in coming years. The responses received are presented as Fig. 1.5
and will be discussed in more detail in the following sections of this chapter.
1.2 Challenges
Taking the above responses, the key issues can be summarised as:
• Design
• Privacy and Security
Fig. 1.4 Mobile phone use (after [13])
• Complexity and Ethics

• Ageing Population
• Users
• Sustainability
• Education
Each of which will be briefly discussed in the following sections.
1.2.1 Design
Conventional approaches to engineering design typically follow a path such as

that defined by the simplified V-Model of Fig. 1.6 with integration being achieved
through a structured system definition followed by a process of system develop-
ment supported by appropriate testing regimes to support verification and valida-
tion. Individual modules and sub-modules, including those from external sources,
are then tied into the design by a process of specification, test, verification and
validation to ensure overall system functionality.
This approach has evolved over many years through the synergetic interaction
between design theory and design practice. However, it is the case that design the-
ory must inevitably lag behind practice where the possibilities afforded by new
Fig. 1.5 Practitioner
responses regarding
challenges facing
mechatronics
Fig. 1.6 Simplified V-Model
technologies are being explored, perhaps without necessarily a full understanding

of their capability or implications.
In the case of the Cyber-Physical Systems and the IoT, the system is a dynamic
entity which smart objects, and hence users, enter or leave depending on con-
text and need. In the majority of instances, the cloud-based components will be
unknown to the user prior to their being adopted for use, and the same may well
apply to any functional smart objects. This leaves the designer with the issue of
ensuring that the system is not vulnerable to their inclusion, while recognising the
ability of the system to self-configure as required.
Essentially, therefore the user specifies system function and content after which
the system autonomously configures selecting required software and data com-
ponents from cloud with information then becoming a commodity whose value
is determined by user context. Where physical components are involved, as for
instance in a smart home environment, identification and selection will be by the
user with guidance. A challenge for designers is to provide tools to enable the
implications of the dynamic system configuration to be explored at the earliest
stages of the design process, and to integrate these outcomes into device function-
ality as appropriate [18].
1.2.2 Privacy and Security
Many of the devices associated with the IoT have the capacity to gather large vol-
umes of personal data, much of which may be held in areas and ways unknown
to the user. This data is then subject to the possibility of analysis, with associated
risks of misinterpretation impacting on privacy [20–23]. However, this must be
balanced against the potential ability to extract beneficial knowledge, particularly
Table 1.2 Perceived threats Threat Probability (%)

to system security (after [19])
Data leakage 17
Employee error 16
Employee-owned device incidents 13
Cloud computing 11
Cyber attacks 7
Disgruntled employee 5
External hacking 5
All of the above 19
None of the above 8
within the context of IoT-based applications such as eHealth [24]. In the wider
context of security, the ability of systems to protect themselves against intru-
sion is of increasing importance, both at the personal and the corporate level.
Table 1.2 shows the perceived levels of threat based on a survey conducted by the
Information Systems Audit and Control Association [19].
It is therefore clear that there is an increasing burden on system designers to
place privacy at the core of their design process within the context each of the
Internet of Things, Cyber-Physical Systems and Big Data, and that this must be
reflected in the design process itself and the methods and tools to support this.
1.2.3 Complexity and Ethics
As systems become increasingly complex and begin to operate with greater auton-
omy, issues are raised regarding the ability of all stakeholders to understand their
nature and function across a range of applications and environments from health-
care to autonomous vehicles [25–28]. This is particularly the case where responsi-
bility for the wellbeing, or indeed the life, of an individual or individuals is being
entrusted to the system [29]. Other issues include:
• Dual-use of technology—Technologies such as drones can be associated with
beneficial applications, as for instance in crop management, but also for military
and other purposes.
• Impact of a technology on the environment—The introduction of technologies
into an environment can disrupt and change that environment in a variety of
ways, even when the underlying intent is benign.
• Impact of technology on the global distribution of wealth—The use of technolo-
gies can increase the separation between differing societal groups, even within
the same country [13].
• The digital divide and the associated socio-technological gap—There is an
increasing separation between the ability to access and use the services pro-
vided through the cloud.
Fig. 1.7 An ageing population in Europe (after [32])
• Ensuring fair access to technologies—Controlling access to technology can act

as a restriction on development.
• Technological addiction—Individuals becoming addicted to the technologies
that they use [30].
• Technological lock-in—Individuals can become locked into specific technolo-
gies, a simple example being the choice between Apple and Android.
• Dehumanisation of humans and anthropomorphism—By taking away responsi-
bility for their activities and wellbeing [31].
1.2.4 Ageing Population
Faced with an ageing population, Fig. 1.7 shows the past and predicted changes in
the distribution of age groups within Europe,1 questions are raised as to how best
to use technology to support the elderly, and to try to provide them with increasing
levels of independence in old age. In particular, there is a need to ensure appropri-
ate levels of mobility within both the physical and information domains to prevent
individuals retain independence and engagement with society [33, 34].
1.2.5 Users
As has been seen, the availability of Internet-capable devices has had a signifi-
cant impact on social behaviour through the use of social media, but also allows a
much more ready access to information than has historically been the case. Such
devices also support increased levels of interaction with the environment, as for
instance in the case of a smart home. Additionally, the introduction of wearable
1Similar data can be found for other global regions.

devices provides opportunities for developments in areas such as eHealth and

mHealth to support individual wellbeing [35], in turn raising issues of privacy and
the control of personal data.
However, there is also a need to develop new forms of user interface to support
a wider range of users in their ability to interact with such systems. In particular,
there is an increasing requirement to be able to capture user intent and context in a
way which does not require complex forms of communication or knowledge about
the underlying technology.
1.2.6 Sustainability
There is a recognised need to move towards more sustainable forms of society-cen-

tred around the individual and their needs and structured around the effective man-
agement and use of all available resources as suggested by Fig. 1.8. In the context
of mechatronics [36, 37], this integrates into concepts such as those of the smart
home and the smart city where information is used to manage daily activities.
For instance, it is estimated that, on average, finding a parking place in a
German city requires about 4.5 km of driving which for a vehicle emitting around
140 g of CO2/km will generate at least 630 g of unnecessary CO2, and signifi-
cantly more in stop-and-go traffic. By linking knowledge of available parking
spaces with the vehicle destination through appropriate communications, much of
this excess could be eliminated [38]. Other sustainability issues impacting on cit-
ies are suggested in Table 1.3.
Overall, therefore there is a move towards the creation of sustainable socie-
ties with individuals and the core to address issues such as ageing populations,
resource availability and management, climate change and resilience [40–44].
Referring to mechatronics and the Internet of Things, an underlying requirement
Fig. 1.8 Sustainability domains

Table 1.3 City problems (after [39])

US/Canada Europe Asia Latin Africa
America
Average population (millions) 1.4 2.5 9.4 4.6 3.9
Population density (per km2) 3100 3900 8200 4500 4600
Water consumption (litres per capita per 587 288 278 264 187
day)
Water loss rate (%) 13 23 22 35 30
CO2 emissions per capita (tonnes) 14.5 5.2 4.6 No data No
data
Waste volume (kg per capita per year) No data 511 375 465 408
in achieving sustainability is the effective management and use of all resources;

technical, physical and human, through the integrated use of information serviced
by a range of smart objects.
This in turn implies the effective and appropriate use of information to support
the engagement of individuals in all aspects of their lifestyle through the adoption
of novel and innovative approaches to understanding, structuring and managing
the physical and information environments, and the relationships between them, as
part of a knowledge economy configured around the Internet of Things. Consider
two different urban scenarios as follows:
Scenario 1: New Build—The aim is to achieve integration of the physical and
information environments from the outset, supported by access to facilities such
as high-speed broadband networks and the ability to deploy a full range of smart
technologies within those environments.
Scenario 2: Established Communities—These represent the majority of the popu-
lation and means that the introduction of infrastructure changes will need to take
account of the impact on the existing environment, and the adaptation of that envi-
ronment to the needs of technology.
1.2.7 Education
Mechatronics education has always faced the challenge of balancing appropriate

levels of technical content with the understanding of the requirements for integra-
tion across the core disciplines of mechanical engineering, electronics and infor-
mation technology [16, 17, 45–47]. Given the growth in the technological base
over the last 40 plus years as suggested by Fig. 1.9 [17], the challenge facing
mechatronics course designers in achieving that balance has become significantly
more complex.
Fig. 1.9 Development and diversification of mechatronics technologies (after [17])
Fig. 1.10 The challenges of innovation

In addition to the challenges to course design associated with developments in

technology a number of other factors need to be taken into account. These include:
• Changes in delivery
– Massive On-line Open Courses (MOOCS) [48].
– Tutorial and workshop based learning support.
– Blended learning [49].
– Impact of social media on learning [50].
• Structural Issues
– Distributed learning resources.
– Time value of content.
– Collaborative working.
A key element for the future is therefore that of encouraging an innovative
approach to mechatronics through education (Fig. 1.10).
1.3 Chapter Structure
The book is structured around a series of chapters from invited authors, each of
whom is an expert in a particular area of mechatronics. In each case, the authors
were challenged to establish the current state of the art using their own research or
professional expertise as the starting point and then to try to isolate and identify
those key areas in which significant development is needed or likely to take place
in coming years. The chapters themselves are organised as set out in Table 1.4.
1.4 Summary
Though the core technologies and concepts remain essentially unchanged, the
nature of what constitutes mechatronics has changed significantly since the concept
was originally proposed, and that change is likely to continued at an accelerating
rate. Some of the issues and challenges be addressed have been identified in the
preceding sections, and will be developed and expanded in subsequent chapters.
Table 1.4 Chapter structure Chapter(s) Subject area

1 Introduction
2 and 3 Issues and Challenges
4–8 System Design, Modelling and Simulation
9 Manufacturing Technology
10–12 Internet of Things and Cyber-Physical Systems
13 Communication and Information Technologies
14 and 15 Mechatronics Education
16 Conclusions
Acknowledgements The authors would like to acknowledge the input made over many years
by a multitude of colleagues, researchers and students to the background, structure and rationale
of both this chapter and the book. There are far too many of you to name individually, but our
thanks to you all!
References
1. Kozono H, Yokota S, Mori T (2008) Mechatronics Coined by Tetsuro Mori. In: 12th
International conference mechatronics technology (ICMT 2008)
2. Shi J, Wan J, Yan H, Suo H (2011) A survey of cyber physical systems. In: 2011 International
conference on wireless communications and signal processing (WCSP), pp 1–6
3. Poovendran R (2010) Cyber-physical systems: close encounters between two parallel worlds.
In: Proc IEEE 98(8):1363–1366
4. Sha L, Gopalakrishnan S, Liu X, Wang Q (2009) Cyber-physical systems: a new frontier. In:
Machine learning in cyber trust. Springer, Berlin, pp 3–13
5. Kortuem G, Kawsar F, Fitton D, Sundramoorthy V (2010) Smart objects as building blocks
for the Internet of Things. IEEE Internet Comput. 14:44–51
6. Zorki M, Gluhak A, Lange S, Bassi A (2010) From today’s Intranet of things to a future
Internet of Things: a wireless and mobility-related view. IEEE Wireless Commun 17:44–51
7. Atzori L, Iera A, Morabito G (2010) The Internet of Things: a survey. Comput Netw
54:2787–2805
8. Gubbi J, Buyya R, Marusic S, Palaniswami M (2013) Internet of Things (IoT): a vision,
architectural elements, and future directions. Future Gener Comput Syst 29:1645–1660
9. Mayer-Schönberger V, Cukie K (2013) Big data: a revolution that will transform how we
live, work, and think. John Murray, London
10. Chen Min, Mao Shiwen, Liu Yunhao (2014) Big data: a survey. Mobile Netw Appl
19(2):171–209
11. Hilbert M, López P (2011) The world’s technological capacity to store, communicate, and
compute information. Science 332(6025):60–65
12. Hesham A-J (2013) The mobile shift: how connected devices are changing consumer atti-
tudes and behaviour. comScore, Reston
13. Smith A, Page D (2015) US Smartphone use in 2015. PewResearchCenter
14. Smith A (2010) Americans and their gadgets. Pew Research Center
15. Zickuhr K (2011) Generations and their gadgets. Pew Research Center
16. Bradley D (2010) Mechatronics—more questions than answers. Mechatronics 20:827–841
17. Bradley RD, Ferguson I, Isaacs J, MacLeod A, White R (2015) The Internet of Things—the
future or the end of mechatronics. Mechatronics 27:57–74
18. Lee AL (2008) Cyber physical systems: design challenges. In: 11th IEEE international sym-
posium object oriented real-time distributed computing (ISORC), pp 363–369
19. Information Systems Audit & Control Association 2012 @ www.isaca.org/Pages/2012-
Governance-of-Enterprise-IT-GEIT-Survey.aspx. Accessed 6 Nov 2015
20. FTC Staff Report (2013) Internet of Things: privacy and security in a connected world. @
www.ftc.gov/system/files/documents/reports/federal-trade-commission-staff-report-novem-
ber-2013-workshop-entitled-internet-things-privacy/150127iotrpt.pdf. Accessed 10 Oct 2015
21. Kambatla K, Kollias G, Kumar V, Grama A (2014) Trends in big data analytics. J Parallel
Distrib Comput 74(7):2561–2573
22. Jing Q, Vasilakos AV, Wan J, Lu J, Qiu D (2014) Security of the Internet of Things: perspec-
tives and challenges. Wireless News 20:2481–2501
23. Weber RH (2010) Internet of Things—new security and privacy issues. Comput Law Secur
Rev 26:23–30
24. Raghupathi W, Raghupathi V (2014) Big data analytics in healthcare: promise and poten-
tial. Health Inf Sci Syst 2(3) @ www.biomedcentral.com/content/pdf/2047-2501-2-3.pdf.
Accessed 17 Oct 2015
25. Gonz CA (2014) Teaching ethics in engineering and design, the necessity of concurrent engi-
neering. In: Proceedings of 16th international conference engineering and product design,
University of Twente, pp 700–705
26. Mateosian R (2013) Ethics of big data. IEEE Comput Society, Los Alamitos, pp 60–61
27. Goodall NJ (2014) Machine ethics and automated vehicles. In: Road vehicle automation.
Springer, Berlin, pp 93–102
28. Lilley D, Wilson GT (2013) Integrating ethics into design for sustainable behaviour. J Des
Res 11(3):278–299
29. Bonnefon J-F, Shariff A, Rahwan I (2015) Autonomous vehicles need experimental ethics:
are we ready for utilitarian cars? @ http://arxiv.org/pdf/1510.03346v1.pdf. Accessed 6 Nov
2015
30. Kuss DJ, Griffiths MD, Binder JF (2013) Internet addiction in students: Prevalence and risk
factors. Comput Hum Behav 29(3):959–966
31. Kile F (2013) Artificial intelligence and society: a furtive transformation. AI Soc
28(1):107–115
32. EuroStat @ http://ec.europa.eu/eurostat/statistics-explained/index.php/Population_structure_
and_ageing#Population_structure. Accessed 6 Nov 2015
33. Domingo MC (2012) An overview of the Internet of Things for people with disabilities. J
Netw Comput Appl 35(2):584–596
34. Boulos MNK, Al-Shorbaji NM (2014) On the Internet of Things, smart cities and the
WHO healthy cities. Int J Health Geographics 13(10) @ www.biomedcentral.com/
content/pdf/1476-072X-13-10.pdf. Accessed 6 Nov 2015
35. Portet F, Vacher M, Golanski C, Roux C, Meillon B (2013) Design and evaluation of a
smart home voice interface for the elderly: acceptability and objection aspects. Pers Ubiquit
Comput 17(1):127–144
36. Millbank J, Bradley DA (2010) Sustainability issues within mechatronics: Pt I—issues and
context, mechatronics 2010, Zurich
37. Millbank J, Bradley DA (2010) Sustainability issues within mechatronics: Pt II: technologi-
cal impact, mechatronics 2010, Zurich
38. Siemens Pictures of the Future @ www.siemens.com/innovation/en/home/pictures-of-the-
future/infrastructure-and-finance/livable-and-sustainable-cities-advanced-parking-manage-
ment.html. Accessed 6 Nov 2015
39. Siemens Pictures of the Future @ www.siemens.com/innovation/en/home/pictures-of-the-
future/infrastructure-and-finance/smart-cities-facts-and-forecasts.html. Accessed 7 Nov 2015
40. Royal Academy of Engineering (2011) Engineering the future @ www.raeng.org.uk/publica-
tions/reports/engineering-the-future. Accessed 7 Nov 2015
41. Royal Academy of Engineering (2012) Smart infrastructure: the future @ www.raeng.org.uk/
publications/reports/smart-infrastructure-the-future. Accessed 7 Nov 2015
42. Epstein MJ, Roy M-J (2001) Sustainability in action: identifying and measuring the key per-
formance drivers. Long Range Plan 34:585–604
43. Chappells H, Shove E (2005) Debating the future of comfort: environmental sustainability,
energy consumption and the indoor environment. Building Res Innov 33(1):32–40
44. Vanegas JA (2003) Road map and principles for built environment sustainability. Environ Sci
Technol 37:5363–5372
45. Kortuem G, Bandara AK, Smith, Richards M, Petre M (2013) Educating the Internet of
Things generation. IEEE Computer Society, pp 53–61
46. Yilmaz Ö, Koray T (2011) A mixed learning approach in mechatronics education. IEEE
Trans Educ 54(2):294–301
47. Jovanovic V, Verma AK, Tomovic M (2013) Development of courses in mechatronics and
mechatronic system design within the mechanical engineering technology program. In:
11th Latin American and caribbean conference on engineering and technology (LACCEI),
pp 14–16 @ www.researchgate.net/profile/Vukica_Jovanovic/publication/259497303_
Development_of_Courses_in_Mechatronics_and_Mechatronic_System_Design_within_
the_Mechanical_Engineering_Technology_Program/links/0c96052c4e6b99abcb000000.pdf.
Accessed 7 Nov 2015
48. Liyanagunawardena TR, Adams AA, Williams SA (2013) MOOCs: a systematic study of the
published literature 2008–2012. Int Rev Res Open Distrib Learn 14(3):202–227
49. Owston R, York D, Murtha S (2013) Student perceptions and achievement in a university
blended learning strategic initiative. Internet Higher Educ 18:38–46
50. Benson V, Morgan S (Eds) (2014) Cutting-edge technologies and social media use in higher
education. In: Advances in higher education and professional development. IGI Global
Chapter 2
Mechatronics Disrupted
Maarten Steinbuch
2.1 How It Started
The field of mechatronics started in the 1970s when mechanical systems needed
more accurate controlled motions. This forced both industry and academia to
explore sensors, and electronic assisted feedback, while using mostly electri-
cal drives instead of, for instance, mechanical cam shafts in production facilities.
This introduction of feedback-controlled motion formed the basis for the need to
enable mechanical engineers and electronic engineers to work better together and
to understand each others language. Note that in those days control engineering
departments were mostly part of the electrical development or research depart-
ments of industry and academia. Various initiatives were also undertaken to
develop a common language or methodology. Some institutes pushed mechatron-
ics forward as being a new discipline.
In industry, the design teams were typically forced to really discuss at the spec-
ification level deeper insights from within their specific disciplinary knowledge.
Computer-assisted design and simulation tools really boosted the field in the late
1980s and 1990. An example of the project-oriented mechatronics way of working
has been the development of optical storage devices such as that of Fig. 2.1 [1].
Teams of mechanical designers, using their finite element programs, and electron-
ics and control specialists, with their specific simulation tools, codeveloped mech-
anisms with very tight specifications on manufacturability, cost and dynamics.
In that same time frame of the 1980s, in many industries and academia,
mechanical engineers started more and more to also address dynamics and con-
trol, and control groups started to emerge also in mechanical engineering depart-
ments, all of which signalled a move away from the mono-disciplinary approaches
of Fig. 2.2 [1].
M. Steinbuch (*)
Eindhoven University of Technology, Eindhoven, The Netherlands

DOI 10.1007/978-3-319-32156-1_2
18 M. Steinbuch
Fig. 2.1 An optical storage device with a balanced rotating arm by philips electronics NV
Fig. 2.2 Many mono-disciplinary solutions for a given problem [1]
2.2 Computer Controlled Devices
The rapid development of the personal computer, enabled the better use of simula-
tion and design tools, and hence improving the overall design process and quality
of exchange of design ideas in an early phase. However, and equally relevant, the
PC-enabled digitized computer controlled mechatronic systems testing and imple-
mentation. This required addressing the role of computer science engineering and
showed the need to include the software discipline, but to a still rather limited
extent. This also led to include more and more the field of systems engineering as
2 Mechatronics Disrupted 19
a way of working in industry on more complex products and high tech systems.
However, thinking about the ‘common’ language, or at least to understand each
other better, clearly is far less trivial between the hardware and software domains,
than within the hardware domain itself.
From a research perspective, the questions start at the discrete time level, i.e. how
to the use the computer to implement control functions such that the performance
previously done with analogue implementation was maintained as much as possi-
ble. However, soon the higher level supervisory control modes were taken into the
mechatronics field, and this forced research to make the switch towards the much
more difficult questions of discrete event systems, facing continuous time dynamics
in the mechanical system. This has led to the research field of hybrid systems within
the systems and control discipline. This part forms the natural interface between the
hardware (the ‘old’ mechatronics) and the software (computer science) field.
2.3 Applications
The performance improvements due to mechatronic thinking have been profound

and are broadly acknowledged. Applications of mechatronics can be found in
many products and production environments. Although in the early days, the con-
trol of electric motors was an often seen application, mechatronic thinking also is
used in the design of hydraulic systems, piezo driving actuators, the modelling and
control of production equipment, scientific equipment, opto-mechatronics, auto-
motive mechatronics, etc.
Overseeing the inflow of submitted mechatronics papers over the last few years,
more application papers are submitted on medical devices, on high precision sys-
tems, drones (UAV), automotive and robotics. The papers on scientific achieve-
ments on modelling languages and tools have reduced, meaning probably that
appropriate tooling is now more common. The same seems to be true for papers
on education in mechatronics. This was a hot topic in the late 1990s, where good
examples were found including experimental work for the students.
There are not so many discussion papers anymore about what could be called
the mechatronic design method, because it is by now maybe clear that part of the
innovation done in mechatronics in practice has more to do with helping disci-
plines to communicate, preferably via the use of shared models or quantified
simulations. The scientific methods addressed in mechatronic journal submission
are mostly seen in the systems and control area, where the mechatronic applica-
tion is often used as a validation or simply as a show case. An emerging field is
the use of optimisation algorithms, not only for finding optimal control laws, but
more and more also for component design, up to system topology optimisation as
a new design tool [2]. The core of the mechatronic submissions and community
still is Mechanical Engineering, Electrical Engineering and the area of Systems
and Control. The interrelation with Computer Science and Physics is still rather
limited, but this is going to shift to coming years.
20 M. Steinbuch
2.4 Multi-physics
High-end mechatronic systems such as wafer scanners such as that of Fig. 2.3 for
optical lithography or electron scanning probes and in space applications and sci-
entific instrumentation, have an error budget that is getting closer to being a flat
distribution over the various sources.
For instance, for modern wafer scanners thermal and cooling-fluids-induced
vibrations now are as significant as mechanical modal vibrations excited by the
actuators. This has to do with the extreme conditions and requirements; moving
an 80 kg mass with accelerations more than 10 g, and achieving accuracies below
nanometres with mKelvin temperature variation [3]. This means that the ‘normal’
mechatronics and its motion control systems now start to have a dynamic inter-
action with the thermal and fluid control dynamics. The overall performance
assessment and design improvements now start to cover not only mechanical and
electrical/electronic and software disciplines, but also physics issues like ther-
mal and fluid partial differential equation-based modelling. And what will be the
impact for mechatronics design thinking when we include the possibilities of addi-
tive manufacturing? If a 3D industrial metal or ceramics printer can be used to
freely shape our mechanisms, how to arrive at an overall optimal design?
Fig. 2.3 Wafer scanner

Fig. 2.4 The performance

complexity (resources) trade-
off [4]
The performance trade-off can now only be lifted to the next level if we are
able to handle this complexity by proper systems engineering and the inclusion of
more disciplines.
In Fig. 2.4 this trend is depicted in the form of a performance versus resources
plot. Resources could be money, people, development time, computer power,
energy, etc. The performance typically is accuracy, throughput and robustness/
reliability. The curve shows that achieving more performance does cost more and
more resources, until not feasible. In the figure, examples are also plotted; first,
a simple transmission gear system, having low performance (in terms of accu-
racy) and also requiring limited resources. The second, example in the figure is a
modern wafer scanner as the example of extreme performance and needing huge
resources.
The curve implies that in order to further boost innovation, we need to incor-
porate two means. First, by addressing all relevant disciplines, so including for
instance physics, we will be able to increase performance. Second, by introducing
a systems engineering approach we can handle complexity in a better way, and
hence, go left on the resources axis.
2.5 Robotics
Almost opposite to the high-end systems as described above, the robotics field also
influences the mechatronics area. Here, it is not the multi-physics discipline that
is required, but the computer science field to cope with unstructured and chang-
ing environments. In robotics, the developments are directed towards vision, map-
ping, and localization, so understanding the environment (‘world modelling’) but
also the field of Artificial Intelligence (AI)—which has already been a promise
for decades, but could evolve rapidly in coming years. Both areas are currently
in an accelerating phase because of the upcoming autonomous vehicles. The dis-
ruption seen in the automotive industry is huge, both in the area of power trains
22 M. Steinbuch
Fig. 2.5 The Preceyes eye surgery robot [5]
(i.e. electric drives and transmissions), and the use of computer science, as for
instance the sensors in a modern car, including the rapid developments in autono-
mous functions implemented in passenger cars as well as in commercial vehicles.
This in fact is all about mechatronics, AI, controls!
The field of robotics, including autonomous cars, could be treated as a sepa-
rate research area, next to mechatronics, but for instance the speed requirements
of industrial robots or the accuracy requirements of surgical robots such as
the Preceyes robot of Fig. 2.5 necessitate the inclusion of the description of the
dynamic behaviour of the robots. The change from rigid body modelling towards
flexible systems, then directly makes it in the heart of mechatronics. The same
holds for the systems engineering thinking and the system topology optimi-
zation, which is also similar in hybrid power trains for vehicles. So where does
mechatronics end and robotics start?
2.6 Cyber-Physical Systems, Smart Industry

and the Internet of Things
The shift from decentralized mechatronic systems towards networked con-

nected systems is known as the field of cyber physical systems, referring to
the field of cybernetics. The research questions are how to guarantee stability
and performance during or after packet (information) loss, and how to deal
with variable delays. The domain is even further away from the hardware of
mechatronics, but is developing so rapidly, that we should ask the question how
to embrace to potential of network-controlled systems, for instance in the field
of remote condition monitoring and servicing. In the next decade, the explosion
of the Internet of Things (IoT) further necessitates finding the answers to this
question [6].
One application where mechatronics will meet IoT is in the future of our manu-
facturing. The Industry 4.0 or Smart Industry attention is about networked modern
industrial automation.
• What does it mean for the flow of goods through a manufacturing plant if
knowledge of the logistics is shared, if the performance of one workstation is
optimized as part of the total logistics or operation, if service and repair in a
production facility is robust because workstations are flexible and can adapt?
• What does this imply for the industrial robotics and smart mechatronic produc-
tion devices?
• How will this impact the design requirement of our mechatronic devices and
products?
The Internet of Things will not only change the modern factory. It is estimated
that in 2020, 50 billion devices will be connected to internet. This means it will be
entering our households and equipment used at home, as well as our cars. When
wearable electronics are pushed further, and we are surrounded by sensors, we
only need the step towards actuation to be able to closed the loop and by that enter
the world of mechatronics again [6]!
2.7 Towards Systems Integration
Overseeing these developments we could question what mechatronics actually is

or will be. Is mechatronics being disrupted? Has it evaporated already into sys-
tems engineering, is it part of the supporting disciplines, does it enlarge to be the
backbone of cyber physics? Moreover, if biological systems are also going to have
technical devices implemented (Internet of Humans), what is then the role of the
mechatronics discipline? How should we educate people in mechatronics think-
ing, how small or how broad? In Fig. 2.6 the role of systems engineering is used
to enable the necessary integration of the disciplinary as well as the technological
contributions.
In this book many of the mentioned developments will be addressed. We will
not have definite answers for the future of mechatronics, nor for its education, but
we learn also that this should be robust and adaptable because we cannot predict
the future! We know for sure that the pace of technological development is accel-
erating, hence, so should we!
24 M. Steinbuch
Fig. 2.6 Systems engineering integration of disciplines and technologies [4]
References
1. Munnig Schmidt R, Schitter G, van Eijk J (2011) The design of high performance mechatron-
ics. Delft University Press, The Netherlands
2. Silvas E, Hofman T, Serebrenik A, Steinbuch M (2015) Functional and cost-based automatic
generator for hybrid vehicles topologies. IEEE/ASME Trans Mechatron 20(4):1561–1572
3. Butler H (2011) Position control in lithographic equipment: an enabler for current-day chip
manufacturing. IEEE Control Syst Mag 31(5):28–47
4. Steinbuch M, Pahnke K, High Tech Systems Centre. Eindhoven University of Technology.
http://www.tue.nl/htsc
5. Meenink HCM (2011) Vitreo-retinal eye surgery robot: sustainable precision. PhD thesis,
Eindhoven University of Technology
6. Bradley D, Russell D, Ferguson I, Isaacs J, MacLeod A, White R (2015) The internet of
things—the future or the end of mechatronics. Mechatronics 27:57–74
Chapter 3
Future Challenges in Mechatronics
Nicolas Albarello, Alexandre Arnold and Marc Budinger
3.1 Introduction
In the aerospace industry, mastering the design of mechatronic systems is a major

requirement. Indeed, a big part of programme cost is spent on the design of these
systems, which also represent a big part of the product performance. In this
chapter, some of the main challenges that industries will face in coming years in
the field of mechatronics are exposed. These challenges deal with the design of
mechatronic systems, their verification/validation and their operation.
3.2 Challenges in Design—Architecture and Sizing
3.2.1 Using Models to Size an Architecture
The physical architecture of embedded technological systems such as the electro-

mechanical actuators of flight control systems, Fig. 3.1a [1] or power electronic
modules of supply network, Fig. 3.2b [2], are an association of components from
different technologies.
Work by Van der Auweraer et al. [3] and Hehenberger et al. [4] highlights that
the design of such multi-domain systems requires different modelling layers as
represented in Fig. 3.2:
• A mechatronic layer has to take into account the functional and physical cou-
pling between components. This level of modelling is usually done using 0D/1D
N. Albarello (*) · A. Arnold

Airbus Group, Toulouse, France
M. Budinger
INSA Toulouse, ICA, University of Toulouse, Toulouse, France

DOI 10.1007/978-3-319-32156-1_3
26 N. Albarello et al.
Fig. 3.1 Multi-domain architecture of an embedded system
Fig. 3.2 Hierarchical design models (after Hehenberger et al. [4])
models [3], also called lumped parameter models, represented by algebraic

equations, ordinary differential equations (ODE) or differential algebraic equa-
tions (DAE).
• A specific domain layer to describe the performance limits and parameters is
necessary in the previous layer, based on a geometric representation. The spe-
cific domain phenomena are generally represented through partial differen-
tial equations (PDE). This level of modelling can be achieved for simplified
geometries using analytical models or, for complex 2D and 3D geometries, for
instance by using numerical approximations such as finite element methods
(FEM).
The design of the power element with a system integrator’s point of view should
allow for the optimizing of the size and specification of components from multiple
technologies interacting together. This system-level design, distinct from compo-
nent design, needs to represent in the mechatronic layer the key information of the
specific domain layer with dedicated models [5–7].
Referred to as “estimation models” by Budinger et al. [7], they enable the
designer to readily take multiple design constraints into account. The models
3 Future Challenges in Mechatronics 27
directly and explicitly link a few primary characteristics, such as overall dimen-
sions of components, to the secondary characteristics needed for the sizing [5]
and optimization [1]. The capacities required of these estimation models are as
follows:
• To present a form that is simple to handle and to implement in different calcula-
tion tools.
• To lend themselves to easy analytical manipulations.
• To be reusable in an area slightly different from the one where they were ini-
tially employed.
To satisfy these constraints, simplified analytical models are often used. Among
these, scaling laws have proved effective in representing a physical phenomenon
over wide ranges of variation [6]. However, these models are only valid under cer-
tain conditions, among which one can mention geometry and material similarities
and uniqueness of the driving physical phenomenon.
For the system designer the models should be as predictive as possible.
Detailed finite element models, able to precisely predict the physical phenomena,
are still much time-consuming in such a context. Despite a recent thrust of work
on model order reduction, the computational cost of finite element models remains
prohibitive in the preliminary design phase. The use of meta-modelling techniques
[7, 8] is thus interesting for this purpose. A challenge for mechatronic design is the
development of meta-modelling techniques specifically dedicated to the selection
of components of system from an integrator point of view. A paper by Budinger
et al. [7] proposes a meta-modelling method based on scaling laws which extract
simple, global expressions of estimation models from local numerical simulations
(FEM).
3.2.2 Coupled Disciplines for the Design of Mechatronic

Systems
The design of systems as those in Fig. 3.1 is driven by the following main aspects
to meet the various requirements: integration (mass, geometrical envelope)
between airframe and actuated load, resistance to environment (thermal and vibra-
tion), instant power and energy saving, dynamic performance, service life, reli-
ability, resistance to or tolerance of failures. Table 3.1 summarizes these different
design viewpoints and the possible associated modelling levels for a model-based
design.
These multiple design viewpoints generate real challenges when optimizing
such systems. To take account of these criteria in the same loop, tools coming
from multidisciplinary design optimization (MDO) can be useful [8]. MDO is a
field of engineering that uses statistical and optimization methods to solve design
problems incorporating all relevant disciplines represented by 2D/3D FEM
Table 3.1 Design views and associated modelling levels during design of mechatronics systems,
an example of a flight control actuator
Requirements Corresponding Algebraic Differential CAD (3D) FEM (3D)
design or sizing models (0D) algebraic (1D)
viewpoints equations
Integration Mass ⊗ ⊗
Geometrical ⊗ ⊗
envelope
Mechanical Transient stress ⊗ ⊗
resistance Fatigue/Thermal/ ⊗ ⊗ ⊗
Wear stress
Vibration ⊗ ⊗ ⊗
Reliability Life time/MTBF/ ⊗ ⊗
Failure rate
Failure/Critical ⊗ ⊗
cases: winding
short circuit, jam-
ming, shock
Dynamic/ Natural modes ⊗ ⊗ ⊗
Control Bandwidth ⊗
Precision ⊗ ⊗
Power/Energy Transient input ⊗ ⊗
power
Energy ⊗ ⊗
consumption
analyses, 0D/1D simulations and algebraic calculations simultaneously. Each

specific computation is considered as a black box which can be called directly,
analysed with design of experiments (DoE). All calculations can be linked
together and used for design exploration, sensitivity, optimization and robustness
analyses; iSight [9], Optimus [10] and ModelCenter [11] are examples of such
frameworks.
Optimization tasks require a small computational time for the models. The sur-
rogate models or meta-models [8], simplified or approximate descriptive model
of another model, can be used for representation of specific domain layer mod-
els (FEM) into the mechatronic layer. The mechatronics layer generally treats
dynamic systems in the time domain and uses transient time simulation but meth-
ods relying on approximating the time domain behaviour by evaluating the domi-
nant harmonics of the load profile [12] can be attractive during optimization of the
design.
These optimization and statistical enabling tools allow the steps of the design
process to be linked in a freely chosen sequence. They do not, however, pro-
vide help in choosing how the connections in the calculations are structured, or
which parameters are to be taken into account as design parameters, constraints or
objectives.
Knowledge-based engineering (KBE), a discipline which combines knowledge-

based approaches and computer-aided design, can also be useful for design sup-
port. KBE software packages are dedicated to centralizing all the know-how and
expertise for the design of a specific product. Scientific concepts and methodolo-
gies used in these environments are as follows:
• Knowledge bases and computer-aided design: the knowledge of the components
is capitalized with un-oriented algebraic equations (declarative approach) [13].
These components can be easily assembled to describe different architectures.
• Constraint networks, graph theory and optimization: the set of equations ini-
tially defined in declarative form is oriented in order to obtain a calculation pro-
cedure usable by an optimization algorithm.
The adaptation or analysis of the equations may be supported by symbolic compu-
tations, interval calculations or artificial intelligence. These KBE tools can be:
• Linked to a CAD software as in Genus Designer [14] which captures the con-
figuration rules and performs process automation for Solidworks;
• Dedicated to a specific domain as in Enventive [15] for the conceptual design of
planar mechanisms (optimization, tolerance, sensitivity analysis);
• Developed over several domains, as in the case for FST institute software
(TUHH University) which, from the same basis, supports the preliminary
design of lift mechanisms [16], hydraulic networks [17] and EMA of aircraft;
• General, such as TKSolver [18], Ascend [19], Cades [20], Design 43 [21]
or PaceLab suite [22] and are often declarative language that enables a set of
algebraic equations to be used with different inputs depending on the design
objectives.
These tools can provide interesting and important help for the designer especially
when the system becomes complex with multiple technologies. Paired with MDO
tools and dedicated meta-modelling techniques, they might represent the future of
the design of mechatronic systems.
3.2.3 Ability to Synthesize Optimal Architectures
When designing a mechatronic system, numerous solutions can be envisaged at

the architecture level. The task of selecting the most appropriate architecture is a
complex task that is currently mainly done by manually defining, assessing and
comparing envisaged architectures. The use of design synthesis and optimiza-
tion techniques at the architecture level permits the designer to envisage a broader
range of solutions, among which are potentially innovative ones, and to compare
them on a formal basis (using well-defined metrics) in order to select the most
appropriate one.
Engineering design synthesis [23] is a set of techniques that leads to the synthe-
sis of engineering artefacts (2D/3D shapes, architectures, etc) based on knowledge
about the purpose of the artefact, its expected properties and design knowledge
(explicitly formalized or extracted from prior designs).
Optimization techniques iteratively modify some initial solutions (generally
randomly generated) in order to optimize the characteristics of the tested solutions.
However, they barely consider design knowledge in order to generate feasible
solutions.
Coupling design synthesis and optimization permits generation of feasible solu-
tions and the finding of the most performing ones. It is generally more efficient
than a manual process since the explored design space can be larger and since the
process is not influenced by cognitive biases (e.g. beliefs). However, this requires
an ability to assess any generated solution along with all the defined selection
criteria (optimization objectives/constraints). Examples of the use of this type of
techniques are for the design of robot arms [24], vacuum cleaners [25] or aircraft
cockpits [26].
3.2.4 Study of Safety and Availability in Mechatronic

Systems
Reliability, Availability, Maintainability and Safety (RAMS) criteria are often

part of the study carried out during the design process. A number of such studies
must be performed for each considered alternatives in order to quantify the per-
formances of the different architectures with respect to these criteria (reliability,
availability, maintainability, etc).
Currently, these studies are carried out by specialists who build RAMS mod-
els of the mechatronic system and run analyses on them to generate conclusions.
RAMS models are typically built using dedicated formalisms such as Petri nets,
Bayesian networks, reliability block diagrams or more advanced languages such as
Altarica [27] and Figaro [28].
In order to speed up the studies, it would be an advantage to link RAMS stud-
ies to (descriptive or behavioural) architecture models. Indeed, much information
embedded in design models can be reused during RAMS studies.
A first approach considers enriching design models with RAMS data (failure
modes, reliability rates, etc) in order to be able to automatically generate RAMS
models. For instance, SysML [29] models (with a specific profile) can be used to
automatically compute the system-level failure rates [30].
Another approach considers linking design models to RAMS models in order
to ensure the consistency of the RAMS model. In the frame of the MODRIO pro-
ject [31], a prototype was developed to automatically generate Figaro models from
Modelica [32] models and a Figaro knowledge base [33]. Modelica specific con-
structs are used to declare the correspondence with a Figaro block from the knowl-
edge base and other necessary information.
3.2.5 Functional Virtual Representation of the Product
The design and the integration of mechatronic systems is a multidisciplinary

design process which requires multiple domains to collaborate and exchange
information. In today’s very large companies and in the extended enterprise, mas-
tering these information flows becomes essential for the efficiency of the design
phase.
One way of improving the communication between teams is to share a common
virtual representation of the product that integrates all viewpoints. This representa-
tion permits the different teams to have a view on other team’s constraints, and to
always have access to the latest version of the design. Also, it enables the consid-
eration of impacts from other domains during simulations.
An integrated view would also provide a more robust basis for decision-making
since it would permit to have a view of all constraints and objectives of all con-
cerned disciplines.
The main technical enablers for this type of representation is the capability to
exchange engineering data between different teams working with different tools
and data formats and the capability to integrate models coming from different
disciplines.
3.3 Challenges in Verification and Validation
3.3.1 Ability to Virtually Validate a System
The verification of mechatronic systems is a very expensive task for industrials

(especially in the aerospace industry). Indeed, it requires both the systems to be
available and the development and manufacturing of test benches that are often
not reusable from programme to programme. A means to reduce the cost of test-
ing is virtual testing. This practice aims to develop virtual means (system models
and system simulation environments) to test a system and verify its conformity to
requirements. For instance, tests on real wings can be replaced by finite element
models.
The main benefit of virtual testing is cost, since virtual test benches generally
cost much less than real test benches and are often reusable from one programme
to another. Indeed, many systems are similar from programme to programme
and their models can in general be adapted to meet the new design with a limited
effort.
A less obvious benefit is that virtual test benches permit the stimulation of
the system in conditions that are closer from the real stimulation that the system
will encounter during its lifecycle. For instance, on a wing bending test, loads are
applied locally on the real test bench while they can be applied uniformly on the
virtual one, thus representing real loads in a more accurate way. Some aspects
(e.g. thermal effects) can be also considered more easily in a system model while
they require very expensive means of testing if a real test is envisaged.
To enable virtual testing, several aspects of the verification process must be
well managed. First, of course, the validation of the model and of its simulation
environment must be carried out. This can be done by comparison of results with
test bench data or flight test data. In this aspect, model calibration and uncertainty
management techniques are required.
3.3.2 Formalize Model Requests to Model Suppliers
More and more, models are used to perform verification and validation (V&V)
activities on systems. These models are often designed by the suppliers of the sys-
tem (internal or external customers). However, being able to state what is expected
from the model in terms of functionalities, domain of validity, precision, etc is still
a challenge. Indeed, the requester has a view on the overall simulation environ-
ment (i.e. other interacting models, simulation inputs, etc) that is rarely communi-
cated to the model developer in a formal way. This often leads to several iterations
before the expected model is actually supplied to the model requester.
Recently, a model identity card was proposed as a standard description of
model requests [34]. The MIC permits to describe some of the desired character-
istics of the model in order to guide its development. First, interfaces of the model
must be defined describing ports of the model and exchanged variables. Secondly,
four sections of model information must be filled in as follows:
Object—model name, granularity level, reference documents
Object context usage—time computation, tool
Method—model dimension, method, linearity
Model quality—accuracy, verification, validation
Another potential use of this type of standardized model specification is the reuse
of existing models. Indeed, formalizing the characteristics of a model enables pos-
terior searches in model databases and reuse of models in different contexts.
3.3.3 Ability to Exchange and Seamlessly Integrate Models

Between Industrial Partners
In relation to the objective of building multi-system simulation platforms to verify

mechatronic systems, a major element is the ability to exchange and integrate mod-
els. Currently, model integrators tend to force their suppliers to use one tool and to
follow specific modelling procedures (e.g. AP2633 [35]). On their side, model sup-
pliers would like to use their own tools to model the system. Being able to exchange
and integrate models coming from different tools would thus be off great advance.
The functional mock-up interface (FMI) initiative [36] goes in this direction by
providing a tool-independent standard for the exchange of dynamic models and for
co-simulation. It permits the generation of “neutral-format” models (under the
form of a C-code and xml1 files) that can be seamlessly integrated in compatible
platforms. Currently, around 70 tools are supporting the standard.
The use of these standard model exchange forms provides flexibility since the
constituent models for a simulation platform can be developed in a number of dif-
ferent tools. The model providers are thus free to choose and change their pre-
ferred tool without impacting on the overall simulation framework. On the side
of model integrators, flexibility is also ensured since the integration platform can
be chosen and changed among a set of available tools without impacting existing
models.
This type of standard might also replace a lot of the point-to-point interfaces
between tools that are developed in-house for specific needs or sold by tool vendors.
For companies, it can represent huge savings in development or licensing costs.
3.3.4 Formal Verification of a Mechatronic System Through

Models
Detection of system flaws in the very early design phases has always been at the
core of model-based systems engineering (MBSE) to reduce global development
time while increasing the quality of the final product. To this day, simulation is the
most common approach to verify the behaviour of the system under development.
But there is an inherent major drawback, the limitation to a finite number of test
scenarios.
Formal verification techniques enable proving that a model is indeed compliant
to its specification, even if case scenarios are infinite. Among them, model check-
ing is able to perform verification on a computer in an automated process.
The use of model checking is already a common practice in some high-tech
industrial areas such as aerospace, railways, microcomputers and more generally
in the development of any critical embedded system, to guarantee an optimal relia-
bility. Techniques and tools have evolved to overcome some original limitations of
model checking and today it is possible to handle physical models with continuous
and discrete parts. Known as hybrid model checking, this opens up new applica-
tion perspectives, especially in the field of mechatronics.
In the current state of the art, hybrid model checkers are usually limited to
proving safety properties (i.e. the system will never enter a certain set of states),
because they often rely on over-approximation. This makes them good candidates
to prove the correctness of airplane collision avoidance manoeuvres for instance.
1Extensible mark-up language used to define rules for encoding documents in a format which is
both human-readable and machine-readable.

Fig. 3.3 Verification of a satellite memory using hybrid model checking
Figure 3.3 shows another case example, a satellite has the mission to capture
Earth images upon request and download them to ground stations whenever they
become visible; the goal is to formally verify that the memory buffer of the satel-
lite will never be exceeded, based on a specific ground stations configuration and a
maximal number of requests per orbit. The requests are discrete, whereas the data
transfers are modelled continuously.
Hybrid model checkers differ from one another in their expressivity (e.g. which
kinds of differential equations they support) and the over-approximation methods
they provide.
When proving safety properties is not enough, hybrid theorem provers such as
KeYmaera [37] can be used as another formal verification option in a mechatronic
context. These try to automate the mathematical proof of the requested proper-
ties, but usually require some advanced inputs from the user along the demon-
stration in order to come to a conclusion, which may be very tricky for complex
systems.
3.3.5 Ability to Optimize Test Campaigns
Sometimes the formal verification of a mechatronic system is not feasible, for

instance because of complex modelling artefacts or scalability issues. Possible
or not, test campaigns will still be needed once the product is built to check its
conformity to the specification models and validate it against the requirements. In

any case, the limitation to a finite number of test cases which can be executed (first
in simulation, then in real life) makes it essential to identify the most relevant and
representative ones.
The tendency is to generate optimal test cases automatically. Two main ele-
ments determine how this is done: first the test selection criterion, which defines
what is driving the test case generation and second the test generation technology,
which is the algorithm actually producing the results. Typical examples for both
are given below [38].
Test selection criteria:
• Structural model coverage criteria—these exploit the structure of the model to

select the test cases. They deal with coverage of the control-flow through the
model, based on ideas from control-flow through code.
• Data coverage criteria—the idea is to split the data range into equivalence
classes and select one representative from each class. This partitioning is usu-
ally complemented by the boundary value analysis, where the critical lim-
its of the data ranges or boundaries determined by constraints are additionally
selected.
• Requirements coverage criteria—these aim to cover all the informal system
under test (SuT) requirements. Traceability of the SuT requirements to the sys-
tem or test model/code can support the realization of this criterion. It is targeted
by almost every test approach.
• Test case specifications—when the test engineer defines a test case specification
in some formal notation, these can be used to determine which tests will be gen-
erated. It is explicitly decided which set of test objectives should be covered.
• Random and stochastic criteria—these are mostly applicable to environment
models, because it is the environment that determines the usage patterns of the
SuT. A typical approach is to use a Markov chain to specify the expected SuT
usage profile. Another example is to use a statistical usage model in addition to
the behavioural model of the SUT.
• Fault-based criteria—these rely on knowledge of typically occurring faults,
often designed in the form of a fault model.
Test generation technology:
• Random generation—random generation of tests is done by sampling the input

space of a system.
• Graph search algorithms—dedicated graph search algorithms include node or
arc coverage algorithms such as the Chinese Postman algorithm,2 which covers
each arc at least once.
2Also known as the route inspection problem or postman’s tour.

Fig. 3.4 Model-based testing using explicit test models (top) versus design/specification models
(bottom)
• Model checking—model checking is a technology for verifying or falsifying

properties of a system, but can be used to generate test cases based on the given
counter examples.
• Symbolic execution—the idea of symbolic execution is to run an executable
model not with single input values but with sets of input values instead. These
are represented as constraints. With this practice, symbolic traces are gener-
ated. By instantiation of these traces with concrete values the test cases are
derived.
• Theorem proving—usually theorem provers are used to check the satisfiability
of formulas that directly occur in the models. Here it uses mathematical proce-
dures to search through the possible execution paths of the model so as to find
test cases and counter examples.
• Online/Offline generation technology—with online test generation, algorithms
can react to the actual outputs of the SuT during the test execution. This idea
is used for implementing the reactive tests also. Offline testing means that test
cases are generated before they are run. A set of test cases is generated once and
can be executed many times.
The most appropriate test generation technology often depends on the kind of
source model to deal with. Some approaches create test cases from a test model,
whereas other ones take design models as input, as shown in Fig. 3.4. Both can
be behavioural models, but the first kind describes what the tester does to test
the SuT (i.e. which stimulations he gives and verifications he makes), whereas
the latter describes how the SuT is working. The point of view is thus differ-
ent (tester versus implementer), as well as the objective (validation versus
solution).
3.4 Challenges in Operation
3.4.1 Ability to Use Design Models to Improve Maintenance
Currently, there is little reuse of design models in the maintenance of aerospace

systems. However, this knowledge about the behaviour of systems can be used to
improve diagnosis, prognosis and maintenance planning.
In diagnosis, behavioural models can be used to correlate the observations made
on the system and some failure modes or some degradation of the components. This
permits the estimation of the current state of components (e.g. filter clogging) and the
performance of maintenance operations in a more accurate and faster way since the
root cause of a system failure might be detected without needing to inspect all its parts.
Similarly, prognosis activities try to predict the future state of the system in
order to anticipate failures and plan preventive maintenance tasks. For instance,
estimating the remaining useful life of a system by analysing the data transmitted
by its sensors permits the planning of specific maintenance tasks (e.g. tank refill),
to anticipate needed resources (spare parts) and adapt operations accordingly.
Technically, the use of models for diagnosis/prognosis generally requires state
estimation techniques (e.g. linear regression, Kalman filters). These techniques
permit to minimize the error between the observations on the real system and the
model by playing on some parameters of the model (failure, degradations). Since
several configurations of the model may match the observed behaviour, the defini-
tion of observed variables is the main driver for an efficient diagnosis.
In practice, embedding the model is not always possible because of required
computational resources. However, the diagnosis/prognosis tasks can be done on
ground in light of transmitted telemetry (offline PHM).
3.4.2 Ability to Use Design Models to Improve Control
Another use of design models to improve operations is in their use for the con-
trol of the system. Indeed, in some cases, there is an interest to use the knowledge
contained in behavioural models in the control logic of the system. This is known
as model predictive control (MPC). The use of MPC is particularly adapted if the
system has slow dynamics (e.g. chemical plants) or if the control must consider a
long-term usage of the system (e.g. plan resource usage for a mission).
An example of usage of MPC is for energy management. For instance, a simple
model of a hybrid propulsion vehicle can be used in the power control algorithm
to optimize the fuel burn and the use of batteries given a particular mission. This
enables a significant increase in performance compared to a classical control algo-
rithm. However, the certification of such intelligent algorithms is still a challenge.
Also, since embedded models are constrained by the real-time requirement, and
since design models are generally not designed for such applications, a simplifi-
cation of the models must be achieved. This simplification process introduces a
trade-off between real-time performance and representativity of the model.
3.5 Conclusion
Some of the main challenges in the design of mechatronics system were exposed
from an industrial viewpoint. The main driver for this evolution is the reduction of
development costs and time as well as the improvement of the designed products
in terms of cost and performances.
As can be seen, many of these challenges deal with the virtualization of the
product to improve its design, its validation or its operations. Indeed, virtualization
enables more flexibility in the different stages of the development at lower cost.
In design, the multiplicity of components and of specific domains of
mechatronics systems requires seamless integration of FEM and system-level
models during design. For this purpose future works can focus on dedicated meta-
models for mechatronics sizing activities and easy assembly of models thanks to
graph-based MDO approach.
In V&V, future work should focus on the formal verification of mechatronic
systems since it would considerably lower the costs of certification.
References
1. Budinger M, Reysset A, Halabi TE, Vasiliu C, Mare J-C (2013) Optimal preliminary design
of electromechanical actuators. Proc Inst Mech Eng Part G J Aerosp Eng 228(9):1598–1616
2. Giraud X, Budinger M, Roboam X, Piquet H, Sartor M, Faucher J (2014) Methodologies
for the optimal design of the integrated modular power electronics cabinet. In: European
Conference on More Electric Aircraft
3. van der Auweraer H, Anthonis J, de Bruyne S, Leuridan J (2012) Virtual engineering at work:
the challenges for designing mechatronic products. Eng Comput 29(3):389–408
4. Hehenberger P, Poltschak F, Zeman K, Amrhein W (2010) Hierarchical design models in
the mechatronic product development process of synchronous machines. Mechatronics
20(8):864–875
5. Liscouët J, Budinger M, Maré J-C, Orieux S (2011) Modelling approach for the simulation-
based preliminary design of power transmissions. Mech Mach Theory 46(3):276–289
6. Pahl G (2007) Engineering design: a systematic approach. Springer, Berlin
7. Budinger M, Passieux J-C, Gogu C, Fraj A (2014) Scaling-law-based metamodels for the siz-
ing of mechatronic systems. Mechatronics 24(7):775–787
8. Sobieszczanski-Sobieski J, Haftka R (1997) Multidisciplinary aerospace design optimization:
survey of recent developments. Struct Optimisation 14(1):1–23
9. Dassault Systems Isight webpage www.3ds.com/products-services/simulia/products/isight-
simulia-execution-engine/latest-release/?xtmc=isight&xtcr=1 and www.simulia.com/prod-
ucts/portfolio/. Accessed 10 Nov 2015
10. Noesis Optimus website www.noesissolutions.com/Noesis/. Accessed 10 Nov 2015
11. Phoenix Integration ModelCenter www.phoenix-int.com. Accessed 10 Nov 2015

12. Malmquist D, Frede D, Wikander J (2014) Holistic design methodology for mechatronic sys-
tems. Proc Inst Mech Eng Part I J Syst Control Eng 228(10):741–757
13. Allain L (2003) Capitalisation et traitement des modèles pour la conception en génie élect-
rique, Institut National Polytechnique de Grenoble
14. Genus Designer www.solidworks.com/sw/products/details.htm?productID=2104. Accessed
10 Nov 2015
15. Enventive www.enventive.com. Accessed 10 Nov 2015
16. Pfennig M, Carl UB, Thielecke F (2009) Recent advances towards an integrated and opti-
mized design of high lift actuation systems. SAE Int J Aerosp 3(1):55–64
17. Dunker C, Thielecke F, Homann T (2011) New computer-aided methods for preliminary
architecting and sizing of aircraft hydraulic systems, Deutscher Luft- und Raumfahrtkongress
18. Universal Technical Systems TKSolver www.uts.com. Accessed 10 Nov 2015
19. Ascend www.ascend4.org. Accessed 10 Nov 2015
20. Vesta Solutions Cades www.vesta-system.fr/en/products/vestacades/. Accessed 10 Nov 2015
21. Alber R, Rudolph S (2003) ‘43’—A Generic Approach for Engineering Design Grammars.
In: AAAI Spring Symposium on Computational Synthesis, AAAI Technical Report SS-03-02
22. Pacelab www.pace.de. Accessed 10 Nov 2015
23. Chakrabarti A (ed) (2013) Engineering design synthesis: understanding, approaches and
tools. Springer Science & Business Media, Berlin
24. Chakrabarti A, Bligh TP (2009) An approach to functional synthesis of mechanical design
concepts: theory, applications, and emerging research issues. Artif Intell Eng Des Anal
Manuf 10(4):313
25. Wyatt DF, Wynn DC, Jarrett JP, Clarkson PJ (2012) Supporting product architecture design
using computational design synthesis with network structure constraints. Res Eng Des
23(1):17–52
26. Albarello N (2012) Model-based trade studies in systems architectures design phases, Ecole
Centrale Paris
27. Arnold A, Point G, Griffault A, Rauzy A (1999) The AltaRica formalism for describing con-
current systems. Fundam Informaticae 40(2–3):109–124
28. Bouissou M, Villatte N, Bouhadana H, Bannelier M (1991) Knowledge modelling and reli-
ability processing: presentation of the FIGARO language and associated tools, Electricite de
France (EDF)
29. OMG SysML www.omgsysml.org. Accessed 10 Nov 2015
30. Helle P (2012) Automatic SysML-based safety analysis. In: Proceedings 5th International
Workshop Model Based Architecting and Construction of Embedded Systems
(ACES-MB’12), pp 19–24
31. ITEA3 MODRIO www.itea3.org/project/modrio.html. Accessed 10 Nov 2015
32. Modelica Specification 3.2 Rev. 2 www.modelica.org/news_items/release_of_modelica_3_2_
revision_2. Accessed 10 Nov 2015
33. Bouissou M, de Bossoreille X (2015) From Modelica models to dependability analysis.
IFAC-PapersOnLine 48(7):37–43
34. Sirin G, Paredis CJJ, Yannou B, Coatanea E, Landel E (2015) A model identity card to sup-
port simulation model development process in a collaborative multidisciplinary design envi-
ronment. IEEE Syst J, 1–12
35. Benac C (2003) A380 simulation models. Airbus Stand Dev New Media Support Centre [OL]
36. Blochwitz T, Otter M, Åkesson J (2012) Functional mockup interface 2.0: the standard for
tool independent exchange of simulation models. In: 9th International Modelica Conference
37. Symbolaris KeYmaera www.symbolaris.com/info/KeYmaera.html. Accessed 10 Nov 2015
38. Zander J, Schieferdecker I, Mosterman PJ (2011) Model-based testing for embedded sys-
tems. CRC Press, Boca Raton
Chapter 4
TiV-Model—An Attempt at Breaching
the Industry Adoption Barrier for New
Complex System Design Methodologies
Craig Melville, Xiu-Tian Yan and Lixiang Gu
4.1 Introduction
System design and engineering is fundamental to the creation of the devices and
technologies that have become a large part of our lives. Technology companies,
motor vehicle manufacturers and inventors go through this process to develop all
kinds of luxuries and necessities for everyday life in the twenty-first century.
Often from the perspective of the user, the means as to how their products were
created is not of concern; it works so it does not matter. To the designer, the meth-
ods and methodologies are very important tools in their belt, but some tools are
better suited to the job than others. Current methodologies, such as the Mechatronic
V-Model, provide decision-making knowledge and support to designers and enable
simple platforms as the basis for development.
This information source is important, and it tells the designer what it is that
needs to be known, a crucial component of the process for engineers, especially
when designing complex systems. Research finds novice designers to be only
aware of 35 % of their knowledge needs in the aerospace industry [1], showing
that there is a very high competency barrier associated with complex systems.
This high competency standard is but one of the many difficulties that arise
from complex system design relative to conventional system design, but there are
many more, and researchers and companies will always be interested in looking
for new ways to do things. The interest in developing more efficient and effective
methodologies for the design of complex systems can thus be argued for on eco-
nomic terms alone.
C. Melville · X.-T. Yan (*)

Space Mechatronic Systems Technology Laboratory, University of Strathclyde,
Glasgow, Scotland, UK
L. Gu
Beijing Institute of Astronautical Systems Engineering, Beijing, China

DOI 10.1007/978-3-319-32156-1_4
42 C. Melville et al.
Take for instance the example of BAE Systems, one of the world’s biggest and
most successful developers of complex systems in the form of naval, aerospace
and ground platforms for various functions. With £1.3 billion in revenue in 2014
[2], a small investment in research into the design process improvement even for
tiny reoccurring percentile gains would be a simple choice. Academia is one envi-
ronment in which to study the application of new methods and methodologies,
but as Birkhofer et al. [3] show in their work, methodologies born of academic
research are rarely or reluctantly adopted into practice. The reasons for existence
of these adoption barriers range from the lack of perceived usefulness, bad com-
munication of concepts and absence of “proof of usefulness”.
This chapter will introduce the TiV-Model, a design methodology for complex
system projects that aims to put to rest concerns facing the adoption of the method-
ology into practice. The next section will contain a description of the TiV-Model,
how it was developed and will show how it plans on solving design related issues.
This will be followed by the validation planning of the methodology and future
plans for development concerning predicted future challenges within the industry.
4.1.1 Complex System Design
To understand the difficulties in Complex Systems Engineering (CSE), it is

necessary first to distinguish between regular systems engineering and CSE.
Traditionally, engineering design is considered to be an iterative design process
“concerned with the creation of systems, devices and processes useful to, and
sought by, society.” [4]. In short and in a way, CSE is concerned with the inves-
tigation of the means to the creation of complex mechatronic systems such as
robotic systems. The most common understanding of this is the engineering design
process, a general term used to express the series of steps involved in the design of
systems. Figure 4.1 shows the design engineering process in simple form.
Design methodologies and the engineering design processes are, for the most
part, interchangeable. Most methodologies have some focal point such as an opti-
mised critical path, DfX1 methods or some other element that provides for more
favourable results in certain areas relative to other methodologies. A core differ-
ence between complex and conventional methodologies, and also a key identifier
of the former, is that conventional projects try to balance out manufacturability
and repeatability with the product quality. Complex projects in contrast tend to
spare no effort in achieving their goal, even with the use of expensive or difficult
manufacturing processes, particularly where the design is a one-off.
1Design for X.
4 TiV-Model—An Attempt at Breaching the Industry Adoption Barrier … 43
Fig. 4.1 Engineering design

process (after [5])
Table 4.1 Conventional, mechatronic and complex projects [6]

Conventional Mechatronic Complex
Volume production High/Very high High-very low Low/Once
Cost per unit Low/Very low Moderate/High High/Very high
Project size Small/Medium Medium/Large Large/Very large
Average quality Low/Very low High Very high
Focus Manufacturing Product Project
Manufacturing style Highly automated Automated/Repetitive Mostly manual
Project management Linear methods Linear methods Non-linear methods
4.1.2 Complex Versus Conventional System Design
The engineering of complex systems comes with additional challenges that need
to be accounted for in the engineering design process. A majority of these differ-
ences stem from the increased scale and complexity of the project. Issues such as
an increased number of parts and manufacturing operations due to the design’s
physical size can be easily accommodated. Processes related differences due to
budgetary and time constraints, such as the reduced accessibility of physical pro-
totyping, will have to be explicitly addressed and made aware to the designer.
Table 4.1 sets out some of the qualitative properties of conventional products,
mechatronic products and complex projects.
Table 4.1 serves to highlight some of the core issues surrounding the comple-
tion of complex projects, namely.
Complex Design Management—The data and physical output of large scale

CSE projects can be overwhelming when compared to traditional systems engi-
neering. Large capacity servers are required to manage the amount of data, but
the data itself is more varied. For example, a CAD model of a satellite design may
contain separate models of the electronics, chassis, fixtures, mechanisms and heat-
ing elements, possibly further divided by subsystem or payload. This added layer
of complexity must be accounted for in the methodology and the management sys-
tem. The sheer volume of files must be tracked and accounted for as well as appro-
priately labelled for use in a group environment.
Complex Knowledge Base—Complex systems are multidisciplinary in nature
and require a firmer grasp of the required knowledge bases. Tolerances are smaller,
requirements more demanding and designs more convoluted than for regu-
lar mechatronic engineering. Documenting and tracking this knowledge is more
important and computer aided tools are essentially mandatory to ensure each team
is up to date with the huge amount of information, such as operating principles
and design specifications. This wider range and expertise of knowledge means that
specialist teams will be more common; allocating these to areas of the project that
need them is an additional planning complication.
Increased Uncertainty and Risk—As with any high budget project, the more
money invested into it, the more money is wasted on failure. The increased complex-
ity, in the form of increased points of failure, tighter tolerances and non-standard
design practices also brings additional uncertainty in both process and design. Hiring
new graduates and novice engineers may be perceived by management as detrimental
to the project as experienced engineers are expected to take the lead and perform a
disproportionate amount of the work. Design teams require more information, skill
and agency to complete the tasks relative to that of conventional systems engineering.
Design Evaluation and Non-Destructive Testing—High budget projects gen-
erally have the freedom, and are encouraged to develop working prototypes to
test and validate the “real-world” behaviour of their design. In large scale CSE
projects, the nature of the design solution is, however, often one that cannot be
wholly prototyped as cost, time and resource constraints prevent this. In a best
case scenario, subsystems or components can be prototyped, but not full systems.
If full systems are to be tested, it would be in the post-fabrication stages, thus non-
destructive testing is the only way to preserve the system integrity. Reliance on
simulation and on-paper calculations can be considered mandatory otherwise.
4.1.3 Methodology Adoption Resistance
Badke-Schaub et al. [7] summarise the perceived issues with new design methods
and methodologies. Figure 4.2 then shows the common industry reasoning for the
lack of integration of new design models and methods.
Performance issues relate to the absence or uncertainty of proof that the meth-
odology will work as intended or produce results. This stems from a lack of vali-
dation on the part of the creator or of follow-up case studies. The presentation of
Fig. 4.2 Industry perspective barriers to methodology adoption
Fig. 4.3 Comprehensive issue matrix for CSE methodology adoption
the methodology refers to the effective communication of information and its clar-
ity. Process relative issues often involve the intra-task efficiency of the model, for
instance the trade-off of time/cost/flexibility.
If the issues from the CSE perspective are combined with the adoption barriers
list, it is possible to effectively create an issue matrix specific to design method-
ologies within complex system engineering industries by adding an extra column
to Fig. 4.2 as in Fig. 4.3.
The matrix of Fig. 4.3 then provides a list of problems that can be solved at the
methodology level and it is to address the TiV-Model that has been created, a CSE
design model that aims for industry adoption by focusing on the issues that com-
monly prevent industry adoption as well as the issues faced by CSE designers.
4.2 TiV-Model
The TiV-Model of Fig. 4.4 is a CSE design methodology that possesses multiple
traits that make it highly beneficial for use in the complex systems industries with
Fig. 4.4 The TiV-model

a focus on spacecraft and satellite development. The development focused on tak-

ing an existing model platform and adjusting certain characteristics.
Categorised Sequential Task Process—By categorising tasks into stages as a
sequential process, much like traditional design processes, the process can be sim-
plified into a step-by-step programme of tasks. At a glance one would think that
this hinders the application of concurrent engineering. However, by using knowl-
edge databases, the general stage tasks can be partitioned into discipline specific
tasks and goals. This will allow either a traditional approach to the design pro-
cess, or the more modern concurrent approach depending on the preference of the
organisation or team. Allowing for both of these approaches ensures the general
flexibility of the model.
Goal Oriented Process—Traditional design methodologies will sometimes
incorporate specific methods as part of the design process; having methods that are
well proven to work for that specific application can be beneficial, but ultimately
means that the overall flexibility of the methodology is compromised. Additionally,
by focusing on the task rather than the short term goal, new designers may not
understand the purpose of doing such a task, leading to a possible chance of fail-
ure. The TiV-Model instead states which deliverables are required at that stage,
while providing possible, but not definite, methods to accomplish the task.
This allows organisations to adopt the model without changing their pre-imple-
mented methods or tools. This also eliminates the risk of the design team perform-
ing a task simply for the sake of performing a task, which can occur if leaders do
not specify the “Why?” behind it. By focusing on what is needed of them as an end
result, designers can understand the process, focus on the output and are still free
to use whichever method preferable to obtain that output. Again, this flexibility
maintains the value of the methodology across industries and applications.
Idealised Requirements for Accurate, Non-Destructive Validation of Design—
Typically, the prototyping stage in system design would be used to validate spe-
cific functions or systems, the cost/time/resource constraints of CSE make this
form of validation much more inaccessible. The reliance on computer aided means
of validation can, however, be accommodated in the methodology by ensuring that
multiple system and manufacturing/assembly models are created.
Discipline specific models to evaluate parameters such as thermal properties,
yield, kinematics and geometric interferences will be integrated into multifaceted
models, designed to simulate the actual environmental and loading conditions
of the model. For example, launch resonance conditions and the effect of rocket
shroud heating can be modelled independently, but running both of these aspects
together will ultimately give a more reliable result. Combine this with iterative ret-
roactive “reality checks” for the simulation data, and the need for prototyping and
destructive testing can be effectively reduced.
Simplified Resource Allocation Recommendations—For certain stages of the
design process, teams will be formed in some capacity, either functional or dis-
ciplinary. The project planners then have to allocate these teams to tasks pertain-
ing to their expertise. Specialists may also be required for temporary contracting
depending on the variety of the in-house design team. To plan for situations like
these, the specialist knowledge types have been categorised into databases and
linked to the stages where this expertise would be required. In doing so, the pro-
ject planned can look ahead at the kind of disciplines required for the project and
hire ahead of time, reducing cost and time.
Communicable and Understandable Language and Processes—One of the most
crucial aspects of the methodology is its ability to be easily understood, time spent
educating team members does not directly add value to the project, so as little as
possible is the ideal. A framework that is easy and quick to learn will be welcomed
by novice and experienced designers alike, as it enables the newer designers to pick
up the slack earlier in the project without being carried by experienced designers.
The “Tiv” component in the model name enables a simple memory trigger to
remember the stage names and general contents at will (QualitaTive, LegislaTive,
etc.). A simple memory game like this can help boost first time retention of model
concepts. By segmenting the tasks, deliverables and databases into a neat column-
row dichotomy it is hoped that the model can retain a visual appearance that aids
recognition of elements and understanding of task/goal flow.
4.2.1 Model Description
The TiV-Model has the essential steps required for any design process, each of
these are labelled with a memorable name pertaining to nature of the stage.
Tables 4.2 and 4.3 then show the type of information being presented.
Performance—The core problems associated with adoption from this perspec-
tive is the lack of study into validation of the methodology and “proven” useful-
ness. The TiV-Model will be built on the provable performance and is currently
Table 4.2 TiV-Model stage descriptions

Stage Description
Investigative User needs, market research, technology research, specification generation
Legislative Planning, mission statement finalisation, contract agreement, qualitative
spec. document
Qualitative Initial design proposals, mechanical/electrical/control concepts, general solution
proposals, ballpark costing
Quantitative COTS component specifications, detailed design, subsystem design, costing,
custom part design, data scanning for 3-D reconstruction of manufactured parts
Evaluative Prototyping, simulation of launch, system performance and manufacturing
facility, final solution decisions, meshing and model reconstruction based
on scanned data
Productive Part creation/buy-in, subsystem assembly and testing, system assembly and
testing, system modifications and tweet based on reconstructed models from
scanned data
Operative Launch, operation, control, maintenance, repair based on 3-D scanned data,
inspection, and disposal
Table 4.3 TiV-model column descriptions

Column Description
System models These are the models that represent the system through CAD,
concept and detail design, including core outputs
Tasks The core methodology, followed by the designers, shows interactive
processes and critical path
Manufacturing/Assembly Models that relate to the state of manufacturing or assembly, these
models are important for outsourced jobs and production planning
Knowledge database Indication of what types of knowledge is needed and at what point
during the project. Makes resource allocation and planning easier
undergoing the verification and validation process. By performing necessary vali-

dation of the methodology through verifiable means and by performing post exper-
iment case studies on the implementation of the methodology, empirical data can
be given to prove the validity and performance of the TiV-Model.
Presentation—Successful communication of a model’s core principles involves
considering the designer’s point of view during the model’s development. TiV-
Model was initially designed with ease-of-use from a designer’s perspective in
mind. Many of the changes from the base version of this model, the 3-column
model, have involved redefining task and timeline taxonomy to “clean up” the pre-
sentable information on the core methodology view [8].
The organisation of presentable information involved;
• Refining the concept of system models as deliverables representative of the sys-
tem overall.
• Refining the concept of Manufacturing/Assembly models as deliverables
that represent the details required for buy-in, manufacturing and assembly of
components.
• Showing the critical task path and the key deliverables for that stage of the
process.
• Splitting the middle stages of the methodology (Qualitative-Evaluative) by
design discipline and showing rough critical path for each.
• Identifying knowledge databases by discipline.
• Displaying when particular knowledge is needed at which stage.
• Displaying key deliverables with suggested methods, maintaining the option for
alternative methods.
The information displayed on the TiV-Model allows the designer to make a
quick and accurate extrapolation of the meaning behind the visuals and the word-
ing. Methods are “advertised” and encouraged, but ultimately subject to change
depending on the approach of the designer or organisation. This flexibility is com-
municated by showing that the task is outcome oriented, with methods paths only
suggested and not enforced. Designers with the most basic systems engineering
knowledge can develop an understanding of the process and a natural experiment
that shows this will be discussed.
Knowledge base taxonomy is divided into general disciplines that are shown in
the model as well as where they are best applied. This ensures planners recognise
where knowledge is to be applied within the project.
Process—The process issues were addressed through changes made by logi-
cal reasoning, the effectiveness will be demonstrated in the experiment referred
to above and discussed later. As already mentioned, flexibility is ensured by goal
orienting the tasks, leaving the method open to the organisation’s preference, yet
offering options and suggestions for placeholder methods. This aids new designers
in making decisions that would otherwise require more information or expertise.
Support from management is an extension of how well integrated the methodology
is from bottom-to-top in the organisation. However, the success of integration is
subject to acceptance at both management and user level. Direct benefits to man-
agement of the project would come from the interactive program planned for the
final development stage of the TiV-Model.
Complex Systems Engineering—With the increased uncertainty associated with
CSE, measures taken in the methodology can offset this. As mentioned before,
by presenting suggestions for methods and clarifying where specific knowledge
should be used, the uncertainty can be minimised and thus the project risk asso-
ciated with that uncertainty reduced. Solving the problem of a high part and file
count for a CSE project would be the responsibility of the management system in
place and this is addressed as an interactive component of the methodology.
Additionally, the entry skill barrier to new engineers can be reduced by rec-
ognising the knowledge gap between them and more experienced engineers, what
knowledge they need and when. TiV-Model, while being goal oriented, makes
suggestions for possible methods to use to accomplish the task. These methods
are optional, and organisations with prior operating principles can implement their
own methods, but in the absence of that knowledge the designer has the capacity
to retain their agency.
4.2.2 Potential Benefits
Designer
• Easy to use and understand current tasks.
• Information needed is provided at the time it is needed.
• Transparency in planning allows greater agency and communication.
• Novice engineers enabled to contribute more.
• Experienced engineers not relied upon too heavily.
• Choice of method, tools and style dependant on designer or organisation.
Project
• Computer-aided validation focus has higher chance of ensuring correctness first
time.
• Concurrent design options may help improve systems integration quality.
• Clear deliverables helps improve error checking and identifying points of

failure.
• Documentation of each stage is part of deliverables required, meaning retroac-
tive checking and changes can be made during the project.
• More means for design validation.
Planning and Management
• Stage and task breakdown is categorised to ease timescale planning and rough
resource allocation.
• Sequential tasks broke down by discipline, allowing for either a traditional or
concurrent engineering approach.
• Knowledge requirements for each stage outlined, allowing plans for specialist
help.
• Planning is transparent and thus easily communicable.
Organisational
• Flexible goal-oriented design means tools and methods need not change.
• Keeping tools and methods means very quick and easy implementation into
organisation.
• Reduce costs by;
– Supporting inexperienced engineers.
– Using computer-aided design validation as opposed to prototypes.
– Retaining in-house tools and methods.
Industry
• Methodology validation breaks down industry barriers for academic model
acceptance.
• Stepping stone example for new, improved design methodologies.
• Hiring of inexperienced engineers will be justifiable, as risk is reduced.
• Non-destructive and computer-aided means of design validation could reduce
project costs across all projects.
• Flexible and modular methodology sections means TiV-Model can be adopted
into any CSE industry, a potential for a standard.
4.3 Methodology Validation
In order for a new methodology to be accepted as a working, feasible alternative, it

must first be scientifically verified and validated. Verification of design methodolo-
gies involves confirming that the internal logic of the methodology is consistent,
validation involves proving that the methodology will provide the desired output
effectively and efficiently.
4.3.1 Validation Methods
In the realm of engineering design methodologies, research into the validation of

models is somewhat rare. Three suitable models were considered for use in the
validation of TiV-Model.
Technology Acceptance Model (TAM)—The TAM was introduced as a means of
validating tools, models and methods from a usability perspective. The model was
designed specifically for the validation of computer systems, but can be expanded
for general use. The TAM focuses on the acceptance of a model by measurement
of the users intentions; perception of use quantifies validity in this sense [9].
Method Evaluation Model (MEM)—MEM is a method that focuses on the vali-
dation of design models and methods for information systems [10]. Validation is
comparable in many ways to TAM, however, the focus was on validation by user
perception in order to obtain projected performance estimates. However, due to the
limited case study evidence supporting this model’s success, and the need for a
more solid evaluation structure, this model was not selected.
Validation Square—The method that was ultimately chosen to validate the
TiV-Model’s experimental data was the Validation Square [11]. This is a model
used specifically to demonstrate the validity of design methods by scrutinising the
method in four key areas, as shown in Fig. 4.5.
The Validation Square was picked due to its suitability to the field; it was cre-
ated for the purpose of evaluating design methods, the conditions for validation
are far more stringent than the other listed models and the Validation square also
goes as far as to validate itself. The model uses both the theory behind the method
and the empirical data achieved from experiments to verify the method’s structure
and validate its performance. The validation is achieved by challenging the meth-
odology with six logical statements that must be proven true.
1. The individual constructs of the method(ology) are valid.
2. The method(ology) construct is internally consistent.
3. The example problems are relevant to the method(ology).
4. The method(ology) is useful to the example problems.
Fig. 4.5 Validation square

5. The usefulness is a result of applying the method(ology).

6. The method(ology) is useful beyond the example problems.
These six statements have been proposed to be proven true across three experi-
ments. Statements 1–3 can be demonstrated in an uncontrolled natural experiment,
showing that the methodology is designed to be used in the systems engineer-
ing context. Statements 4 and 5 can be shown in a controlled group experiment
designed to prove the usefulness of the methodology compared with other suc-
cessful design methodologies. The sixth and final statement is justified in the
Validation Square as a “leap of faith” once the other five statements have been
proven true; the methodology is as good as valid. However, the project aims to go
one step further and use the TiV-Model in an actual CSE scenario, upon which a
case study will be built to prove the methodology’s effectiveness.
4.3.2 Experiment 1—Natural Experiment
The natural experiment is a means of showing that the TiV-Model is capable of

the core function of producing a complex design solution as part of a mechatronic
design project. This will be the first soft implementation of the methodology in
a realistic use environment with the purpose of obtaining usage data from the
respondent. The experiment is rather simple; a fourth year design engineering stu-
dent was tasked with the design of a robotic solution for the automated application
of icing on cakes, this involved a focus on the design of the mechanism but also
included the control and electronics at a conceptual level. Post-project feedback
is obtained from the respondent in the form of a qualitative feedback survey, fol-
lowed up by an informal feedback session, where efforts will be made to obtain
suggestions to improve suitability for CSE and usability of the model.
4.3.3 Experiment 2—Controlled Comparative Study
The second experiment aims to prove the 4th and 5th statements of the Validation
square. The methodology can be proved to be useful to the example problems by
comparing “usefulness” of the TiV-Model to that of existing successful models.
Some questions for this approach are:
• What variables constitute usefulness in a design context?
• What successful methodologies are valid for comparison?
• How can an experiment be designed to extract these variables?
While the specifics of which methodologies to use are being planned, it is likely the
experiment which will take the form of previously published comparative studies. In
a previous study focused on comparing the V-Model with other life cycle develop-
ment tools, comparison extends no further than the literature and logic [12]. In the
experiment it is aimed to demonstrate the hypothesis of these comparisons via con-

trolled environment, where teams of designers will each be using one of three design
methodologies in a performance incentivised CSE project. Effectiveness will involve
the comparison of output design qualities and efficiency will, much like the natural
experiment, focus on qualitative feedback from the designers as users.
To understand effectiveness there needs to be measurable variables generated
by the project that can be compared. The TiV-Model’s ideal competency is that it
is thoroughly validated, so efforts were made to understand the testing parameters
of arguably the most rigorously tested field of all; medicine.
Validation Lessons from Medicine—When it comes to experiment design and
testing standards, few organisations are more stringent than those involved in
medicine. This is perhaps due to the nature and risk associated with the develop-
ment of pharmaceuticals. There may be no testing standards for experiments with
design methodologies, but methodologies used in medicine can be a useful equiva-
lent benchmark. Frey and Dym [13] discuss in great depth how medicine can be
taken as a useful analogy towards the validation of design methods. Analytical
methods for laboratories have to follow set standards such as those of the US
Food and Drug Administration (USFDA), Current Good Manufacturing Practice
(cGMP) and ISO/IEC in order for their methods to be eligible for validation. The
standards include parameters that can be tested that reflect the success of the end
product. These variables are;
• Accuracy • Precision
• Specificity • Limit of detection
• Limit of quantisation • Linearity and range
• Ruggedness • Robustness
By setting an acceptable threshold for these quantifiable values, medical

researchers can determine effective “success” of a treatment or drug and compare
it to other solutions. Design research can learn from this as many of these factors
have equivalents in a design context.
The specifics of such comparisons are still up for debate, but on a “closest
match” standard. The relevant factors can be determined for the evaluation of
design solutions as opposed to medical ones. Table 4.4 provides the context for
this evaluation in the medical domain.
Table 4.4 Relevant success Test element Design context

measures in medicine
Accuracy Satisfy design requirements
compared to design
Precision Repeatedly satisfy design requirements
Specificity Ability to detect failures
Limit of detection Largest acceptable “failures”
Linearity and range Closeness in solution quality
Ruggedness Design “effectiveness”
Robustness Design “quality”
With these new parameters that determine success, based off equivalents in
medicine, it is possible to continue with designing an experiment that will extract
these parameters and enable the evaluation of the methodology to take place.
4.3.4 Experiment 3—Case Study
To prove the sixth statement, and determine that the methodology is indeed fit
for practical use, a CSE project will be undertaken using the TiV-Model as the
methodology of choice. This project will involve the design of a multifunc-
tional mechatronic gripper for fixture on board spacecraft and structures. This
is sufficiently within the intended design area of the methodology as a complex
mechatronic project, demonstrating its original focus. This project will be docu-
mented and examined as a case study, evaluating the success or failure of the pro-
ject based on similar measurable variables as the second experiment. If the project
is successful, the validation will be complete and presentable as proven fact, more
than most academic models can claim.
4.4 Next Steps and Conclusions
4.4.1 Interactive Software Integration
Relating to the goals of increased user-friendliness, the capacity to manage large

scale projects and integrate with an organisation from top-to-bottom, the TiV-
Model will be further developed into a comprehensive methodology and life cycle
management system. By doing so, it is possible to effectively tie methods together
with their respective tools, for example; evaluation of concepts by weighted conver-
gence matrix is meta-linked to a dynamic group shared file that contains a House of
Quality style matrix. This goal is very much inspired by the PLM systems developed
by companies such as AutoDesk, and in heavy use by the likes of BAE Systems.
4.4.2 Closing Remarks
TiV-Model is a proposed solution to the question often asked in design research;

“why are new models slow to come to practical adoption?” It can be shown that
there are various concerns expressed by industry about the suitability of new mod-
els as well as their performance and usability. The key issue, however, is the lack
of proven effectiveness or validation of said methodologies. By ensuring TiV-
Model is thoroughly valid, it can act as an example of breaking down the bar-
rier of acceptance to industry. It also aims to hold its own as a user-friendly and
exible alternative for the CSE industry. Validation of the TiV-Model will use the
fl
Validation Square, a suitably stringent means for design method evaluation, to
prove that it can perform well. This validation process will encompass three exper-
imental steps that mirror the nature of practical use more with each step. By show-
ing TiV-Model can succeed and even thrive in similar projects, it is possible to
remove many of the doubts industry may have about this new academically rooted
model. It also works to satisfy future needs; the need for an overarching set of
tools, methods and methodologies that encompasses CSE is predicted [14]. The
TiV-Model will work towards the goal of a universally compatible architecture to
accommodate new design methods and tools. Alternatively, by providing a verified
and validated foundation, future method development can springboard from TiV-
Model, perhaps even merging as a powerful supplement to the methodology.
Acknowledgement The authors would like to acknowledge and express their thanks for the
financial support for a key international collaboration project entitled Geometric data processing
of large and complex spatial engineering structures—GeomSES, funded by the Chinese Ministry
of Science and Technology.
References
1. Ahmed A, Wallace KM (2003) Understanding the knowledge needs of novice designers in

the aerospace industry. Novice Des Aerosp Ind Des Stud 25(2):155–173
2. BAE Systems (2014) BAE annual report 2014, London. http://investors.baesystems.com/~/
media/Files/B/Bae-Systems-Investor-Relations-V3/Annual%20Reports/bae-annual-report-
2014.pdf. Accessed 2 Nov 2015
3. Birkhofer H, Jänsch J, Kloberdanz H (2005) An extensive and detailed view of the applica-
tion of design methods and methodology in industry. In: Proceedings of 15th international
conference on engineering design (ICED 05), Melbourne, Paper DS35_188.49
4. The Free Dictionary (2015) Engineering design. http://encyclopedia2.thefreedictionary.com/
Engineering+Design. Accessed 2 Nov 2015
5. Teach Engineering (2015) Full design process. www.teachengineering.org/assets/images/Full
DesignProcess_635wTYPOFIXED.png. Accessed 2 Nov 2015
6. Melville C (2012) The complexities and challenges of ship building compared to a domestic
project. PhD thesis, University of Strathclyde and BAE Systems
7. Badke-Schaub P, Daalhuizen J, Roozenburg N (2011) Towards a designer-centred methodol-
ogy—descriptive considerations and prescriptive reflections, the future of design methodol-
ogy. Springer, Berlin, pp 181–197
8. Yan XT, Zante R (2010) Mechatronic design process and its application, mechatronics in
action. Springer, London, pp 55–70
9. Davis FD, Bagozzi RP, Warshaw PR (1989) User acceptance of computer technology: a com-
parison of two theoretical models. Manage Sci 35:982–1003
10. Moody DL (2003) The method evaluation model: a theoretical model for validating informa-
tion systems design methods. In: Proceedings of European conference information systems
(ECIS 2003), Paper 79
11. Pedersen K, Emblemsvåg J, Bailey R, Allen JK, Mistree F (2000) Validating design meth-
ods & research: the validation square. In: ASME design engineering technical conference,
Baltimore. www.researchgate.net/profile/Farrokh_Mistree/publication/236735347_Validating_
Design_Methods__Research_The_Validation_Square/links/0deec519274d06d53b000000.pdf.
Accessed 2 Nov 2015
12. Balaji S, Murugaiyan MS (2012) Waterfall vs V-model vs agile: a comparative study on

SDLC. Intl J Inf Technol Bus Manag 2(1):26–30
13. Frey DD, Dym CL (2006) Validation of design methods: lessons from medicine. Res Eng
Design 17(1):45–57
14. Hastings D (2004) The future of engineering systems: development of engineering leaders.
MIT Engineering Systems Symposium, pp 29–31
Chapter 5
Digital Twin—The Simulation Aspect
Stefan Boschert and Roland Rosen
5.1 Overview
The vision of the Digital Twin itself refers to a comprehensive physical and
functional description of a component, product or system, which includes more
or less all information which could be useful in all—the current and subsequent—
lifecycle phases. In this chapter we focus on the simulation aspects of the Digital
Twin. Today, modelling and simulation is a standard process in system develop-
ment, e.g. to support design tasks or to validate system properties. During opera-
tion and for service first simulation-based solutions are realized for optimized
operations and failure prediction. In this sense, simulation merges the physical and
virtual world in all life cycle phases. Current practice already enables the users
(designer, SW/HW developers, test engineers, operators, maintenance personnel,
etc) to master the complexity of mechatronic systems.
Nevertheless, the technical challenges are further increasing. Software and
especially network connectivity extend the functionality of mechatronic systems.
As the traditional mechatronic disciplines (mechanics, electric and electronics)
are realized in a more integrated way, their interfaces will be more intertwined.
To design these systems and to validate properties by virtual tests in early phases
as well as operation and service support multi-domain and multi-level simula-
tion approaches are necessary. These approaches must be embedded in the system
engineering and development process and reused in all following lifecycle phases.
In the two dimensions of time and level of detail a seamless Digital Twin is
required, which focuses on the following points:
• Reducing time-to-market is today, and will remain so into the future, a key
aspect. The intertwining of simulation models over different levels of detail,
over all involved disciplines and over lifecycle phases must be enforced.
S. Boschert (*) · R. Rosen

Siemens AG, Corporate Technology, Munich, Germany

DOI 10.1007/978-3-319-32156-1_5
60 S. Boschert and R. Rosen
• The Digital Twin should be designed in its principal structure in advance and
needs its own architecture. It extends the pure data which is available in design
and engineering, and is collected during operation and service by simulation
models. These simulation models describe and make available system behav-
iour, performance evaluations and quality considerations. The Digital Twin
provides an interface to different models and data in different granularities and
keeps them consistent.
• Optimization of mechatronic products and systems during their use or opera-
tion, e.g. as an element of production equipment or supply part, will be more
important. The gap between development and operation must be bridged. The
redevelopment of models for operation and service is time-consuming and
cost-intensive.
• The Digital Twin connects different value chains. For example, a Digital Twin
of a product which is used as production equipment in a production system
transports important information (data and executable models) to producers and
manufacturers, e.g. for an easier system integration (virtual commissioning, pro-
duction planning, etc.). This requires that the Digital Twin contains models with
different granularity and needs in general a predefinition of its principal archi-
tecture (structure, content and purposes). The Digital Twin is not a data monster,
which includes everything from all lifecycle phases.
We show in this chapter that simulation has the potential to be integrated into all
phases of system design in the future. As such it will be available as an additional
feature during all operation phases. As more and more features are realized using
software instead of hardware, simulation models are needed to describe all disci-
plines—separately and in combination—and on different levels of detail. In total,
only simulation methods will master the extended challenges coming from new
technologies used in all kind of technical systems.
5.2 The Waves of Simulation in System Development
In the last decades, simulation has developed from a technology largely restricted
to computer experts and mathematicians to a standard tool used daily by engineers
to answer manifold design and engineering questions, as suggested in Fig. 5.1.
In the 1960s–1970s numerical algorithms, generally implemented in
FORTRAN, were used to calculate specific physical phenomena to solve design
problems. However, this was limited to very special cases, as there would have
been few simulation experts. With the increasing spread of workstations and per-
sonal computers the number of users grew rapidly and the provision of simulation
tools for repetitive tasks like control unit design began to be established.
With a higher number of simulation users, the economic growth of tool pro-
viders, technical improvements like increased computing power simulation of
all involved disciplines and on different level of detail was possible. Today,
5 Digital Twin—The Simulation Aspect 61
Fig. 5.1 The Digital Twin is the next wave in simulation technology
simulation is the basis for design decisions, validation and testing not only for
components but also for complete systems in nearly all application fields. This
trend is unbroken and will be continued in the next years.
Simulation models for all aspects can be implemented and will be provided in
parallel with the real component, product or system, and of course, its use does not
stop with the commissioning phase. It will be used, for example, during operation
for optimized operations and during service for lifetime calculation and improved
maintenance.
Other indicators also suggest that the golden time of simulation has just begun.
This can be seen from a look to the evolution of mechatronic systems. In the pre-
liminary note from VDI1 [1] released in 2004, one can find a description of
mechatronics as:
Innovative products require an interdisciplinary combination of mechanical engineering,
electrical engineering and information technology. The term ‘mechatronics’ is the expres-
sion of this
In the last 15 years, the nature of components, products and systems has
changed. In particular, they offer a highly increased number of functions. This
functional extension is realized by a larger number of parts and especially by a
combined use of different disciplines. Beside mechanics, (or more generally phys-
ics) and electricity, the relevance of software and communication, e.g. intra- and
internet connection, is raised. This leads to an increase of complexity and so to a
stronger use of discipline over spanning methods—simulation is a powerful one.
This trend will hold in the future.
Many products which were called “electro-mechanics” some years ago have
evolved to mechatronic products. A very prominent example is a coffee machine
which developed from a simple device for heating water to a fully automatic
1Verein Deutscher Ingenieure (Association of German Engineers).

device with programs for different coffee taste and advanced cleaning processes.
In examples like this, the following three phenomena can be observed:
• Additional functions are offered by the mechatronic product which are realized
more and more by software.
• The number of parts like drives or sensors in components, products and systems
is increased.
• Intensive use of a combination of different disciplines. Software controls via
actuators and sensors the physical section. In many cases the software is exe-
cuted on microcontrollers which are located nearby and not on separated com-
puter units.
These trends will continue, because there are simply no indicators that point to a
change. Quite the contrary, in the near future two other trends will enforce the cur-
rent development:
• Components to systems2—Mechatronic products gain a hierarchical structure or
will be added as ‘elements’ to larger systems.3
• Enforced connectivity4—Such as web technology, internet protocols and
increased computing power in the product or system itself.
Taking a more detailed look into the communication capabilities of future
mechatronic systems, Fig. 5.2 shows the enhanced understanding of mechatronic
systems supplemented by open network capabilities compared with the VDI repre-
sentation [1]. This principle structure was elaborated in the German Industrie 4.0
BMBF project mecPro2 [2]. In this project the term cybertronic5 is used to empha-
size the importance of cyber capabilities for next generation mechatronic systems.
Cybertronic was used as a shorter form for cyber-mechatronic to underline that the
consideration of the physical system remains relevant for the complete system
functionality. This all illustrates that the significance of simulation as a method to
master complexity and to bridge disciplines will further increase.
On the one hand, the requirements for development of future mechatronic sys-
tems are rising, and on the other hand, the evolutionary development of simulation
techniques leads to significant improvement in mastering the complexity during
the design and operation phases. Nowadays in the industrial sector and in techni-
cal companies, simulation is still mostly considered to be a tool for research and
development (R&D) departments. In addition to more traditional calculations,
for instance in multiphysics simulation, aspects like “communication by simula-
tion” and “virtual experience” are emerging applications in various phases of the
2Clarification of terminology—The authors are aware, and readers should be aware, of differing
views and use of terms like elements, systems, products and components.
3This corresponds also to ‘system of systems’ perspectives.
4Many terms are used to express this trend, e.g. Cyber-physical Systems, Internet of Things, Web
of Systems, Industrial Internet.

5The term cyber-physical is more popular, but brings a different understanding of how the physi-
cal basis system is associated.

Communication in open
networks
Machine /
Human
information processing
Cyber-Communication
Capabilities
Human-machine Machine-machine
interface interface
Information
processing
Cybertronic System
Mechatronic System
Actuators Sensors
Basic physical
system
Legend:
Flow of information Flow of energy Flow of material
Fig. 5.2 Enhanced understanding of mechatronic system, as elaborated in the German Industrie

4.0 BMBF project mecPro2 [2]
development processes. Extending simulation to later life cycle phases as a core

product/system functionality, e.g. delivered before the real product itself or sup-
porting the operation by simulation-driven assistance, is the next big trend in
simulation [3]. Therefore, the realization of a “Digital Twin” is the key vision for
industrial application. It includes simplified modelling processes, easy use of sim-
ulation, simulation workflows and seamless simulation along all life cycle phases
including operation support.
5.3 Concept of Twins
The concept of using “twins” dates back to NASA’s Apollo program, where at least
two identical space vehicles were built, allowing the engineers to mirror the condi-
tions of the space vehicle during the mission, the vehicle remaining on earth being
called the twin. The twin was also used extensively for training during flight prepa-
rations. During the flight mission it was used to simulate alternatives on the Earth-
based model, where available flight data were used to mirror the flight conditions
as well as possible, and thus assist the astronauts in orbit in critical situations. In
this sense every kind of prototype which is used to mirror the real operating condi-
tions for simulation of the real-time behaviour, can be seen as a twin.
Another well known example of a “hardware” twin is the “Iron Bird”, a
ground-based engineering tool used in aircraft industries to incorporate, optimize
and validate vital aircraft systems [4]. It is the physical integration of electrical
and hydraulic systems as well as flight controls, with each laid out in relation to
the actual configuration of the aircraft, and all components installed at the same
place as they would be on the real airframe. The actual cockpit for the Iron Bird
is typically displayed by simulators along with a mobile visual system. From this
flight deck, the Iron Bird can be “flown” like a standard aircraft, with a computer
generating the aerodynamic model and environmental conditions such as air den-
sity, air temperature, airspeed and Mach number.
The Iron Bird allows engineers to confirm the characteristics of all system
components as well as to discover any incompatibilities that may require modifi-
cations during early development stages. Additionally, the effects and subsequent
treatment of failures introduced in the systems can be studied in full detail and
recorded for analysis [5].
Due to the increasing power of simulation technologies, and thus more and
more accurate models of the physical components today, the “hardware” parts
in the Iron Bird are replaced by virtual models. This allows system designers to
use the concept of an Iron Bird in earlier development cycles, even when some
physical components are not yet available. Extending this idea further along all
phases of the life cycle leads to a complete digital model of the physical system,
the Digital Twin.
The term Digital Twin was brought to the general public for the first time in
NASA’s integrated technology roadmap under Technology Area 11: Modelling,
Simulation, Information Technology and Processing [6]. In this report the future
development direction of modelling and simulation was outlined:
A Digital Twin is an integrated multiphysics, multiscale simulation of a vehicle or sys-
tem that uses the best available physical models, sensor updates, fleet history, etc., to mir-
ror the life of its corresponding flying twin. The Digital Twin is ultra-realistic and may
consider one or more important and interdependent vehicle systems, including propulsion/
energy storage, avionics, life support, vehicle structure, thermal management/TPS, etc.
Manufacturing anomalies that may affect the vehicle may also be explicitly considered.
In addition to the backbone of high-fidelity physical models, the Digital Twin integrates
sensor data from the vehicle’s on-board Integrated Vehicle Health Management (IVHM)
system, maintenance history and all available historical/fleet data obtained using data min-
ing and text mining. By combining all of this information, the Digital Twin continuously
forecasts the health of the vehicle/system, the remaining useful life and the probability of
mission success. The systems on-board the Digital Twin are also capable of mitigating
damage or degradation by recommending changes in mission profile to increase both the
life span and the probability of mission success [6].
Along with NASA’s ideas, the US Air Force published similar ideas [7] estab-
lishing the Digital Twin is part of the USAF long-term vision addressing a time
frame of the next 30 years. The general idea is that along with every plane a digital
model is delivered, specific to the individual plane. Being “specific to the tail num-
ber”, this digital model includes all deviations from the nominal design. The digi-
tal model will also be flown virtually through the same flight profiles as recorded
for the actual aircraft and the data will be provided by the structural health moni-
toring (SHM) system of the flying plane. Comparing actual sensor readings with
modelling results at critical locations allows engineers to update, calibrate and
validate the model. Changes in the plane configuration like unanticipated damage
will be added to the digital model. The Digital Twin hence always represents the
current state of the actual aircraft. The main application of this digital model is to
determine when and where structural damage is likely to occur and thus to predict
the optimal maintenance intervals.
The aspect of a seamless coverage of the life cycle with digital models is not
mentioned in the above publications. The USAF also introduced the concept of
a Digital Thread [8] for the acquisition of new material to focus on rapid field-
ing, the development, employment and integration of digital design tools across
the acquisition life cycle. The Digital Thread is the creation and use of a digital
surrogate of a material system that allows dynamic, real-time assessment of the
system’s current and future capabilities to inform decisions in the Capability
Planning and Analysis, Preliminary Design, Detailed Design, Manufacturing and
Sustainment acquisition phases.
The digital surrogate is thus a physics-based technical description of the system
resulting from the generation, management and application of data, models and
information from authoritative sources across the system’s life cycle. A Digital
Thread capability is enabled through technical advances in modelling, data stor-
age and analytics, computation and networks. The Digital Thread concept creates
informed decision making at key leverage points in the development process that
have the largest impact on acquisition programs. This would lead to earlier identi-
fication and a broader range of feasible solutions; a structured assessment of cost,
schedule, and performance risk and accelerated analysis, development, test, and
operation.
Looking at the basic message of these descriptions, the definition of the
Digital Twin and the Digital Thread are nearly the same, only differing in posi-
tion during the life cycle. Both concepts make use of all available virtual models
which are interconnected to provide the best possible information. Also, addi-
tional data either from live systems or from historic data are used. Whereas, the
Digital Thread concept is used to support the acquisition phase, and implicitly the
design of new aircraft, the Digital Twin should support its operation and service.
However, as both concepts are based on the same idea using simulation models
to predict the behaviour of the real system, in the following only the term Digital
Twin will be used regardless of the life cycle phase where the concept is used.
5.4 The Digital Twin from the Simulation Viewpoint
The general vision of the Digital Twin refers to a comprehensive physical and
functional description of a component, product or system, which includes more or
less all information, which could be useful in later lifecycle phases. This is from
a technical point of view not feasible. The data volume is too huge, diverse and
totally unstructured. Furthermore, new applications in later phases require specific
preparation of data and information in previous phases. A specific architecture of a
Digital Twin is necessary.
Before we discuss this aspect, we will describe the Digital Twin vision from
a simulation viewpoint. The Digital Twin refers to a description of a component,
product or system by a set of well aligned executable models with the following
characteristics:
• The Digital Twin is the linked collection of the relevant digital artefacts includ-
ing engineering data, operation data and behaviour descriptions via several simu-
lation models. The simulation models making-up the Digital Twin are specific
for their intended use and apply the suitable fidelity for the problem to be solved.
• The Digital Twin evolves along with the real system along the whole life cycle
and integrates the currently available knowledge about it.
• The Digital Twin is not only used to describe the behaviour but also to derive
solutions relevant for the real system, i.e. it provides functionalities for assist
systems to optimize operation and service. Thus, the Digital Twin extends the
concept of model-based systems engineering6 (MBSE) from engineering and
manufacturing to the operation and service phases.
We will discuss four aspects of the Digital Twin.
• Principle approach and benefit
• Architecture of the Digital Twin
• Lifecycle aspects
• Digital Twin and value chains.
5.4.1 Principle Approach and Benefit
In Fig. 5.3 the principle approach of the Digital Twin is shown. Existing IT systems
like PLM, PDM and SCADA7 systems store and provide huge amounts of information
6“Model-based systems engineering (MBSE) is the formalized application of modelling to sup-
port system requirements, design, analysis, verification and validation activities beginning in the
conceptual design phase and continuing throughout development and later life cycle phases” [9].
So a core idea of MBSE is to use digital models to capture interactions of single subsystems and
components at a system level. The system behaviour is tested against these models throughout
the product development process.
7Product lifecycle management (PLM), Product data management (PDM), Supervisory Control
and Data Acquisition (SCADA).

Closing the loop

back to design
Design Engineering Operation Service
phase-specific use bridge over phases phase-specific use
essential Information Digital Twin
digital artefacts
PLM / PDM
Existing IT-Systems Service cloud
SCADA
authoring systems
Fig. 5.3 The Digital Twin uses the essential information originating from different IT systems
and makes it available for succeeding phases
coming from multiple authoring tools and sources, e.g. user requirements, CAD appli-
cations, operation data. The Digital Twin uses this digital information and makes it
available as data and simulation models. Therefore, the Digital Twin includes the rele-
vant data for processing phase-specific simulation tasks. In addition it contains just the
essential information which is required for succeeding steps and phases. In this way the
Digital Twin is smart, increases productivity and leads to new offerings during opera-
tion like assist systems and service applications. Furthermore, the Digital Twin can
close the loop from operation and service back to design of new products or updated
revisions.
Nevertheless, the Digital Twin is a highly dynamic concept growing in com-
plexity along the life cycle based on the application of MBSE concepts. The
Digital Twin is handed over with the product or even before. During operation it is
the basis for simulation-driven assist systems as well as control and service deci-
sions in combinations with smart data approaches. The concept of the Digital Twin
is independent of manifestations or specific realizations.
5.4.2 Architecture of the Digital Twin
The goal of the Digital Twin is to prepare solutions for different but specific objec-
tives and questions. These questions can arise in all lifecycle phases. For instance
in the design phase, if the functionality of different components should be vali-
dated in a simulated interplay. However, the main benefit of the Digital Twin is
expected in later phases. Therefore, the Digital Twin is a product feature, which
is planned from the early stages. Also its primary use is defined and so limited to
solve specific questions.
This requires that a Digital Twin architect describes the purpose(s) of the
Digital Twin. Derived from this goal the tasks are defined, which are later exe-
cuted to answer the questions. The final step is the specification of needed data
and simulation models, which builds the architecture of the Digital Twin for a con-
crete application and set of purposes. A positive side effect of such a well defined
Digital Twin structure is that in some cases new applications—which may not
have been initially thought of—can be realized as well, based on the available (and
persistent) information provided by the Digital Twin architecture.
However, the Digital Twin is still an abstract concept that allows a better com-
ponent, product or system development. The underlying methodology is based on
several aspects. The first is model based development. The information exchange
is no longer focused on documents; instead models are used as a compact means
to exchange information and interdependencies. Another important aspect of the
methodology is the fact that models can be used in different situations as they are
modular and have standardized interfaces (e.g. FMI [10]). A model management
system keeps the single models up to date as changes occur. Also, the model man-
agement system supports the coexistence of different models with different fidelity
and allows choosing the right model for the right application. Choosing the right
model for a “good enough simulation” means that the model with that granularity is
chosen, that is just fine enough to answer the design question, but not finer. Further,
algorithms for the analysis of real-time and historical data are included as well.
5.4.3 Lifecycle Aspects
The relevant parts of the Digital Twin have to be designed in parallel with the
development and physical realization of the observed system. As described above
a Digital Twin simulation architect defines in the very beginning the structure
and interfaces of all simulation models to be contained in the Digital Twin. This
structure combines the single digital artefacts into a comprehensive functional and
physical description. This structure is guided by the intended application fields for
the Digital Twin. Only the relevant models are included and prepared, other digital
artefacts are still contained in the existing IT systems like PLM. As the develop-
ment continues the structure is filled with the real models and associated data. In
the end the Digital Twin becomes part of the physical product. This procedure is
enabled by a consequent use of MBSE techniques.
In this way the Digital Twin is part of the Digital World, see Fig. 5.4, and con-
tains all information and models which are needed to solve tasks in later phases
and to create new values, e.g. assistance systems for operators, user and mainte-
nance personnel. The volume of the Digital Twin will increase during design and
engineering phases. Depending on the specific component, product or system
the transition to the operation or use phase can be realized in different ways. So
it is possible, that not all data and models will be transferred. However, the col-
lected and stored data in the Digital Twin will increase again during operation and
Digital Twin evolves along phases and will be used as an integrated added value during operation/use
Design Engineering Operation Service
Digital World
Digital Twin
Physical System
Fig. 5.4 Digital Twin evolves along phases and will be used as an integrated added value during
operation and service
service phases. A special feature of the Digital Twin is that some of its content will
become part of the real system, for instance an executable simulation model as an
assist system module of the automation software. Thus, the Digital Twin imple-
ments the link-up of parts of the digital world with the physical system.
5.4.4 Digital Twin and Value Chains
We have already stated, that Digital Twins will be elaborated for—at least—com-
ponents, products and systems. If we take a look to value chains, this means that
Digital Twins will overlap in particular at specific points of different value chains.
A good example is a production system. The equipment of a production system
consists of different production units, which are products from other companies.
Digital Twins of these products can be useful for the (virtual) commissioning of
the productions system and also for the operation of the production system, e.g.
for maintenance planning.
From a technical perspective Digital Twins have to skip borders of legal entities
and in many cases to bridge between different proprietary data formats. Similar
challenges and opportunities are visible on the production site. The producer needs
parts and semifinished products and delivers his goods to customers, which use the
product as end customer or for his production.
This leads to the consequence that the Digital Twin has to be modular. This
modularity is used to transfer data and information into other Digital Twins.
Especially in late phases (production, operation) where the Digital Twin has
already gained lots of data lies the main application for the Digital Twins. For
example, product design data can be used for service lifetime calculations and the
product structure can be used to optimize the assembling in production either for
the single components but also for large systems.
The same data and information structures can, in such an environment, exist
in several models in parallel, as the modular structure of the partial models, is not
always employed to the full extent. This has to be decided on a case by case situa-
tion. Each Digital Twin relates the essential part of the existing data and informa-
tion (from existing IT systems) and makes them usable for its specific purpose.
5.5 Use of Digital Twin: Several Aspects During Life Cycle
In this section we show how the change from simple mechanical or mechatronic
components towards mechatronic systems will occur, and how the comprehensive
Digital Twin influences this transition. As an illustrating example, one can think
of an electromotor as a mechatronic component and the combination of the motor,
driving electronics and software as a mechatronic system.
In contrast to current development philosophies dominated by the engineering
of details, in the future the system view and the interconnection between the dif-
ferent development phases will become more important. Simulation will become
an important method for the whole life cycle.
5.5.1 Design Phase
During the concept design of the motor, a number of data and models are created
which lay the foundation of the Digital Twin. The design is mainly determined by
high-level requirements and experience with former developments. This informa-
tion is used for a first abstract design mainly from a functional view of the motor.
Here it is not yet decided, how the concept will be realized (e.g. in software or in
hardware). In further detailing steps, based on the requirements or assumptions,
the design is concretized. During all subsequent phases the current design has to
be validated against the assumptions and requirements. One possible mean for val-
idation is virtual prototypes, which mirror the current development state. As the
development goes on, the virtual mirror image grows as well.
In the example of the electro-mechanic drive system, the basic requirements
and the functional decomposition of the design can be modelled in a structured
way using system description languages like SysML [11]. The dependencies and
mutual interactions of the functions and requirements are shown, which helps
designers to react quickly with a minimum of errors when requirements change.
Many, if not all consequences of the changes can be identified before unneces-
sary detailing is done, especially if automated checking is implemented—which
by the formal nature of the system description is possible even for large complex
systems. In the further detailing of the design—when first decisions are made, by
which the functions are realized, the virtual representation of the design also has to
be detailed.
5.5.2 Engineering Phase
In the next development phase, additional engineering data including executable

simulation models are created. As the engineering questions are now more spe-
cific, the simulation models also have to include further details. There may exist
several different models for the same physical component, depending on the dedi-
cated question they are created to solve. For successful engineering of complex
systems, domain specific models have to be used together with extended system
models, which include the functionality realized in software as well.
In the case of the motor system, many simulation models are created. There are
mechanical models to examine the mechanical stability of the rotor or the motor
mounting. Electrical models are used to calculate magnetic fields and the resulting
forces. Thermal models calculate heat generation due to electrical losses. Other
models describe the electrics and automation, and system models simulate the
whole drivetrain including gears and bearings.
As the complexity of the objects to be designed increases and more and more
disciplines and people are involved, the need for consistency among these models
increases. In an ideal world all information should be stored in a unique place—
not necessarily at the same physical location—to provide a consistent source and
all models should also use this source for all input. This information includes not
only the initial design parameters but also all models which are created during the
development. However, this is often not possible for real systems. But at least a
common awareness among all stakeholders of the development process about
commonly used information is indispensable for a successful development of next
generation mechatronic systems. The Digital Twin as a common repository for all
digital artefacts supports this.
The latest phase of product engineering is the integration of the different com-
ponents to the full system. Here it is necessary that all components fit seamlessly
together. Usually the components are not ready at the same time or are not avail-
able for testing as it is, for example, with the environment where the product (elec-
tromotor) is finally installed. In this case, models from the Digital Twin can be
used to replace physical components in the system integration test for example, as
it is done with the Iron Bird.
Mounting difficulties when parts from external suppliers have to be integrated
can easily detected in advance if a digital representation of the component is pro-
vided along (or in advance) with the physical part. Such a procedure is supported
by dedicated light-weight formats (e.g. JT [12]), that reduce the exchanged infor-
mation to the necessary amount. So the model from the supplier can be integrated
into the model representation from the Digital Twin easily and checked for any
inconsistencies. This procedure works not only for physical components. Software
for automation can also be tested in advance using the virtual representation of the
real system (i.e. virtual commissioning).
Even the final operating environment can be simulated by numerical models to
have a realistic final test before the deployment of the final product.
5.5.3 Model Reuse for Operation
The main purpose of the Digital Twin for mechatronic systems is that the informa-
tion created during design and engineering is also available and ready for eval-
uation during the operation of the system. This is nowadays often neglected, as
design and operation are mainly disconnected life cycle phases from the point of
data usage. An obvious example for model reuse is continuous product improve-
ment, for example, if the intended use of the product changed and product modi-
fications are necessary. In this case the existing models can be used and slightly
modified.
On the other side, if data from operation are also collected systematically as
part of the Digital Twin, they can be used to verify and update the existing models
for real operation conditions such that the gained knowledge can be used for next
generation of products as well. As the Digital Twin is already planned from the
earliest time, the suitable interfaces to interact with real data are already in place.
Thus, the real data can be used as verification input for the simulation models and
lead to their continuous improvement.
Online condition monitoring is also a growing application field for the Digital
Twin. For more and more mechatronic systems, sensors are installed to monitor
the operation and give early warning of a malfunction [13]. However, only rely-
ing on sensor data is sometimes not sufficient. Especially, as particular values may
not be accessible for a direct measurement. In this situation the simulation models
from engineering, provided by the Digital Twin, can be reused after some modi-
fications. The existing models have to be streamlined, to cope with the real-time
data and new requirements from operation. Based on the real sensor data the simu-
lation models extend the measurements towards a “soft sensor”, which can also
acquire virtual sensor data, where real measurements are technically not possible.
By using the simulation models, it is also possible to interpret the measure-
ments in a different way, rather than just detecting deviations from the norm.
Several modes of failure can be simulated for the current situation trying to repro-
duce the actual measurement signals. The comparison of the simulated signals
with measured ones can help to identify the failure mode.
5.5.4 Service Phase
As the Digital Twin provides a smart view on the available system information,
models and results from earlier life cycle phases are accessible also for users of
different disciplines.
During engineering the product has to be designed such that it can withstand
a given load. The proof is done via simulations. However, additional information
from the simulation can be deduced with little extra effort, like the expected life-
time of the part or how “well” the design criteria are fulfilled. Parts which fulfil
these conditions only tightly are natural candidates as causes for malfunction. The
availability of this information allows an improved and enhanced service process,
as the most important (with highest likelihood needed) spare parts are already
known in advance.
Together with data from online measuring, the simulation models and opera-
tion history provided by the Digital Twin are also the base for more flexible ser-
vice planning. Depending on the actual load exposure, the lifetime budget of
the relevant parts—accounted for in the Digital Twin—is deduced. Therefore, a
comprehensive picture exists of the condition of the system which also eases the
inspection planning and spare parts logistics even before a failure occurs.
5.6 Conclusion and Outlook
For mechatronic systems new and novel goals will emerge, e.g. as caused by cyber-
physical systems. These are networked systems, which interact together, and real-
ize new functionalities by cooperation. Aspects like autonomy will be important
and software-driven configuration and use (digitalization aspects) will increase.
Therefore, mechatronics will evolve further. Mechatronic products will gain a more
complex structure and will have more computing power and network connectivity.
This leads to the extended design challenges where simulation will be a key tech-
nology to master it. Simulation will not only become an essential part during the
development of mechatronic systems but it will also become a part of the systems
themselves as well. It will be applicable and executable during the operation of the
mechatronic systems for operation support and new service applications.
The classical goals will still remain valid: time reduction (development time,
time-to-market), adherence to quality and fulfilment of customer needs and
requirements.
Realizing the simulation aspects of the Digital Twin is a key vision from our
point of view to make significant steps forward to reach these goals. It is the smart
way to gather and provide all information stored in existing IT systems which is
essential for life cycle over spanning use. The benefit of the Digital Twin concept
will be the improved consistency, a seamless development process and the pos-
sibility of reuse in later life cycle phases. Further, there is the increased potential
for using the complete set of information in the development of next generation
products.
References
1. VDI-Richtlinie 2206 (2004) Entwicklungsmethodik für mechatronische Systeme

2. BMBF Project mecPro2 (2013). www.mecpro.de. Accessed 2 Nov 2015
3. Rosen R, von Wichert G, Lo G, Bettenhausen K (2015) About the importance of autonomy

and digital twins for the future of manufacturing. In: 15th IFAC symposium on information
control problems in manufacturing, Ottawa, 11–13 May 2015
4. Shafto M, Conroy M, Doyle R, Glaessgen E, Kemp C, LeMoigne J, Wang L (2010) NASA
technology roadmap: DRAFT modeling, simulation, information technology & processing
roadmap technology area 11, Nov 2010
5. Airbus Industries (2015) Innovation. www.airbus.com/innovation/proven-concepts/in-design/
iron-bird/. Accessed 2 Nov 2015
6. Shafto M, Conroy M, Doyle R, Glaessgen E, Kemp C, LeMoigne J, Wang L (2012) NASA
technology roadmap: modeling, simulation, information technology & processing roadmap
technology area 11, Apr 2012
7. Air Force Research Laboratory (2011) Condition-based maintenance plus structural integrity
(CBM + SI) & the airframe digital twin. 88ABW-2011-1428
8. US Air Force (2013) Global horizons: United States air force global science and technology
vision AF/ST TR 13-0, July 2013
9. INCOSE (2007) Model based systems engineering. www.mbse.gfse.de. Accessed 2 Nov 2015
10. Modelisar Association (2014) Functional mock-up interface. www.fmi-standard.org. Accessed
2 Nov 2015
11. OMG, OMG SysML v 1.3 (2015). www.omg.org/spec/SysML/1.3/. Accessed 2 Nov 2015
12. Carter M, Huang J, Sashank Ganti J (2014) JT file format reference version 10.0, Siemens
PLM Software
13. Deckers J (2014) Dem Defekt einen Schritt voraus: Drehmoment-Schwingungsüberwachung
vermeidet Getriebeschäden, bevor sie entstehen, VDI-Fachtagung Schwingungsüberwachung
Chapter 6
Design Processes of Mechatronic Systems
Matthieu Bricogne, Julien Le Duigou and Benoît Eynard
6.1 Introduction
Nowadays, in order to design innovative and multidisciplinary products such as

mechatronics systems, the product development process needs to be rethought or at
least adapted. Current processes are based on organizational models, business
standards IT solutions and so on and have to take into consideration that electronics
and software represent an ever increasing part of the final product [1], as illustrated
in Fig. 6.1. In this context, traditional mechanical engineering, electrical/electronic
and software methods cannot just be coordinated, and common interdisciplinary
design approaches have to be proposed. These approaches will be influenced by
several trends such as agile design methods [2], servitization1 of products [3] and
increasing demand for mass-personalized products [4].
This chapter is divided into two main sections. The first reviews some of the
current design processes of mechatronic systems in order to point out the heter-
ogeneity of practises and the gap remaining to integrate these discipline specific
practises. The second section then presents several industrial trends which will,
from our point of view, greatly influence the future of mechatronic systems design.
1Servitization (also found as servicization or servification), refers to a paradigm of transition of a
product centric offer to a combined product-service offer, underpinning a change of the business
model for the company.
M. Bricogne · J. Le Duigou · B. Eynard (*)

Sorbonne Universités, Université de Technologie de Compiègne, CNRS,
UMR 7337 Roberval, Centre de Recherche Royallieu, CS 60 319,
60 203 Compiègne Cedex, France

DOI 10.1007/978-3-319-32156-1_6
76 M. Bricogne et al.
100%
90%
80%
70%
Functions
Mechanical
60%
50%
40%
30% Electronic
20%
Software
10%
0%
1970 1980 1990 2000 2010 2020
Fig. 6.1 Function assignment over time in the development process of mechatronic systems
(adapted from [1])
6.2 Current Design Processes of Mechatronic Systems
Mechatronic systems have evolved from electromechanical systems with discrete

electrical and mechanical parts to integrated electronic–mechanical systems with
sensors, actuators, and digital microelectronics driven by sophisticated software
modules [5]. This evolution is driven by changes in market conditions, expressed
as new demands such as integrating more functions into the product, reducing the
weight and size of the product, increasing reliability and so on.
Despite these deep evolutions, the main development process, called here
the new development product (NPD) process, is not much changed over time.
Although new requirements for transversal communication between all partici-
pants from different disciplines, engineering activity and cultures are becoming
increasingly important, the NPD main steps have largely remained the same. The
next section will present some of these NPD process models.
6.2.1 New Models of Product Development Process

and Mechatronic Systems Specificities: Focus
on Multidisciplinary Integration
Models of product development process have generally emerged from the mechan-
ical communities [6–8]. Despite this historic connection, the titles of these mod-
els and the step names illustrate the fact that the originators want to present these
process models as generic, whatever the type of product developed, or the type
of technology used. They generally speak of “engineering design” and not of
“mechanical design”. Here, only one product development process model that
is presented in Fig. 6.2 [7] has been chosen as it is an questionable standard, but
a more general summary is also presented by Howard et al. [9]. Based on this
model, specificities of mechatronics systems design are described in this section.
6 Design Processes of Mechatronic Systems 77
Fig. 6.2 Steps in the planning and design process [7]

Needs Clarification and Conceptual Design Phases [9]

Design of a mechatronic system requires multidisciplinary collaboration. To deal
with such multidisciplinary design issues, systems engineering has been proposed
as an interdisciplinary approach [10]. In the mechatronic context, systems engi-
neering is mostly used to consolidate a requirements list, to establish function
structures and to propose preliminary product architecture. For instance, SysML
[11] and associated methodologies [12] are often presented as a solution to support
the need establishment phase whereas model-based systems engineering (MBSE)
activities [13] and Design Structure Matrix [14] are used to support system archi-
tecture elaboration.
Embodiment Design and Detailed Design Phases [9]
During the embodiment design and detailed design phases, one of the biggest
remaining challenges is the data management. For these phases, most of scientific
efforts in the mechatronics field concern product model dedicated to mechatronics
(Core Product Model, MOKA, etc.) [15] to store information, unique bill of mate-
rial (BOM) to federate expert contributions (Fig. 6.3), standards creation/usage to
support data exchange [15, 16], consistency management between expert’s models
[17] and so forth.
This focus on the product data management is mainly due to the fact that meth-
odologies remain specific for each discipline. Some of these discipline specific
methods are presented in the next section.
6.2.2 Models for Mechatronic Development Process
As seen in the previous sections, a strong feature of the design process of

mechatronic system is that it requires a multidisciplinary and holistic development
process. Despite this fact, few specialized models of mechatronic development
process are available in the literature. For several decades, different models, such
as the Waterfall or cascade model [18], the spiral model [19, 20] or the V-model
[21] have been proposed to support the design systems. This type of model can
indeed support the design process at a very macroscopic level, but it does not
support collaboration between designers from different disciplines. In particular,
the multidisciplinary integration, such as hardware–software integration, is not
supported.
Some specialized design process models for mechatronics have been exposed
and compared [15]. The most well known of these processes, the V-Model [22],
presents a general flow for the product development process. To understand how
this process is implemented by industrials, Aca et al. [23] detailed the way the
tasks are dispatched between the different disciplines according to project man-
agement principles, Fig. 6.4 illustrates this situation. Teams are then able to work
concurrently on each sub projects and the multidisciplinary integration is treated at
the late stage of the process.
Fig. 6.3 Unique bill of material for the whole mechatronics system (adapted from PTC: mas-
tering change and configuration management for business advantage (2013). www.ptc.com.
Accessed 20 November 2015)
Fig. 6.4 Mechatronic system development process based on discipline division [23]
In these first sections, development processes main characteristics for E/E and
software engineering have been presented and positioned relatively to NPD gen-
eral process. The heterogeneity of these processes has been underlined to demon-
strate and illustrate the remaining challenges for mechatronic system design and
integration. In the next section, some of the current scientific approaches attempt-
ing to improve multidisciplinary integration are presented in order to illustrate the
remaining work in this field.
6.2.3 Further Challenges for Integration of Mechatronic

Systems: Current Research Approaches
In the previous section, classical mechatronic design processes were exposed and
appear to be very sequential and discipline specific. But developing mechatronic
systems requires intensive collaboration between engineers from different domains
[24, 25] . Therefore, there is a need for concurrent engineering approaches with an
integrated strategy.
The main challenges in design of mechatronic systems [26] can be summarized as
• Exchange of design models and data;
• Cooperative work and communication among the design engineers;
• Multidisciplinary modelling;
• Simultaneous consideration of design from different disciplines;

• Early testing and verification;
• Persistence of a sequential design process;
• Need for tools and methods supporting multidisciplinary design;
• Support of the design of control software.
Some design methods seems to be more adequate for the design of mechatronic
systems (Systems Engineering, Agile Methods, MBSE, etc) and are adapted in
recent works specifically for this purpose.
Systems engineering has been proposed as a multidisciplinary approach to
enable the realization of complex systems [27] apply extensions for every design
phase based on the VDI guideline 2206 in order to improve the development of
systems controlled by a programmable logical controller. The Requirement–
Functional–Logical–Physical (RFLP) approach is a specific V-Model derived
method, particularly adapted to mechatronic systems design and formed of four
phases (requirement, functional, logical, physical) which are each supported by
different technical tools aiding the designers [27]. Zheng et al. [28] propose a two
level process based on V-Model and hierarchical modelling, focusing on inter-
faces to improve multidisciplinary integration in detail design. System engineering
improves cooperative work and communication among the design engineers at the
early stages of the design process, multidisciplinary modelling, early testing and
verification and provides a set of tools and methods supporting multidisciplinary
design. Nevertheless, the detailed design is still very domain specific and needs
further developments.
Agile methods came from software development and are mainly driven by
“accelerate delivery” realized by focusing on small steps. Contrary to established
methods as V-Model based on stage-gate with strict processes and heavy docu-
mentation, agile methods principles are based on interactions between individu-
als, customer collaboration and working products (or prototypes) [29]. But agile
methods in mechatronics systems development is less explored than in software
development. However, few examples exist, as [30, 31] that present agile methods
applied on industrial cases. Stelzmann [32] discus appropriate context to the appli-
cation of agile methods in system engineering context. In this context, Bricogne
et al. [33] propose a framework including engineering actions management, trans-
disciplines activities management, data exchange and collaboration to support
agile methods applied to mechatronic systems design.
Agile Methods improve cooperative work and communication among the
design engineers, multidisciplinary modelling, simultaneous consideration of
design from different disciplines, and early testing and verification. Nevertheless,
large-scale systems, traceability and safety and distributed environment imply the
impossibility to use agile methods, but the agile principles (welcome changes,
self-organizing teams, etc) may still be applicable.
MBSE has attracted numerous researchers’ attention recently. It is consid-
ered as a significant methodology for the design of mechatronic systems with
increasing complexity [34]. The Object Management Group’s Unified Modelling
Language and System Modelling Language (SysML) has been recently widely
used to support the MBSE [35]. Some extensions have been developed for SysML
to support the specific requirements of design of mechatronic systems, such as
automatic simulation [36], design making process [37] etc. MBSE improves
exchange of design models and data, multidisciplinary modelling, simultaneous
consideration of design from different disciplines, tools and methods supporting
multidisciplinary design and support of the design of control software. However,
current studies on MBSE mainly focus on early design phases and seldom involve
the detailed design phases. Moreover, MBSE are more adapted to complex sys-
tems but it is still missing effective and user friendly tools for such development.
Mechatronic systems design new processes are widely explored by academics
and industrials but none of them overpass all challenges, especially in the detailed
design phases.
6.3 Future Trends for the Design of Mechatronic Systems
In the previous sections, some current scientific approaches attempting to address

organizational challenges for mechatronic system design have been presented. In
the future, new challenges will have to be taken into consideration. These chal-
lenges will relate to several trends, such as the ever increasing software element in
products (Fig. 6.1) and the necessary collaboration between systems.
6.3.1 Hardware/Software Design Methods Convergence:

Cyber-Physical Systems Versus Mechatronic Systems
The science of Cybernetics has given birth to a type of system referred to as a

cyber-physical system (CPS). These CPSs refer, according to Rajkumar et al. [38],
to a next generation of systems that require close integration of information, com-
munication and control technologies to achieve stability, performance, reliability,
robustness and efficiency in the management of physical systems of many applica-
tion areas [38]. Although CPSs development brings up specific issues and chal-
lenges, a large majority of integration problems encountered are similar to those
identified in the design of systems involving multidisciplinary teams as required
for mechatronic systems. Despite this observation, the mechatronics and CPS
communities at present seem to only slightly interact and share their expertise. The
explanation proposed here is that mechatronic products are historically electrome-
chanical products introducing more and more electronics and software while CPSs
are historically the “Cyber-Systems” [39], which are more and more interacting
with physical reality, Fig. 6.5 shows the predominance of the software aspect of
this vision.
Fig. 6.5 Cyber systems actors’ vision on cyber-physical systems [39]
Fig. 6.6 Co-influence of mechatronic system and CPS
As can be seen, mechatronics and CPSs challenges and research issues are
complementary. While mechatronics is particularly focused on the hardware part
of a system, CPSs remains focused on the software parts. Nevertheless, these two
areas share a common goal which is to design and produce integrated systems.
Figure 6.6 summarizes this convergence between CPS and mechatronic systems.
6.3.2 The Role of Mechatronics in the Internet of Things

Trend
This convergence between mechatronic systems and CPSs is also subject to the
pressure of the Internet of Things (IoT) trend, “The connection of physical things
to the Internet makes it possible to access remote sensor data and to control the
physical world from a distance” [40]. Thanks to this definition, we can understand
the future role of mechatronic systems into this (r)evolution. IoT is based on smart
objects, which can be considered as a mechatronic system connected to the Internet.
The need for new devices or for the integration of sensors or/and actuators in
new generation of existing products will probably greatly increase the develop-
ment of the mechatronics field. Whereas mechatronics’ application was sometimes
considered for specific and complex systems, this pervasive deployment of smart
objects [40] will greatly amplify the need for mechatronic systems. However,
some authors argue that “there will need to be significant changes to the way
mechatronic, and related, systems are designed and configured” [41] to be fully
integrated in the IoT trend.
They underline that “increasing complexity while managing the transfer of
functionality, particularly from the mechanical domain to the information technol-
ogy and electronics domains, has long been an issue facing system designers” [41,
42]. This state will force “practitioners and educators to further review the ways in
which mechatronic systems and components are perceived, designed and manufac-
tured. In particular, the role of mechatronic smart objects as part of an IoT based
system in which the structure is defined by context is resulting in an increased and
increasing emphasis on issues such as machine ethics, user interaction, complexity
and context as well as with issues of data and individual security” [41].
From a technical point of view, if every mechatronic system can be individually
identified and can conform to a standard protocol, the interoperability of the sys-
tems will increase and make systems even more autonomous and intelligent. New
applications can then be envisioned. Another perspective relies on new synergistic
services offered by mechatronic products compared to an isolated embedded sys-
tem. Both these new opportunities are detailed in the next sections.
6.3.3 The Role of Mechatronic Systems in the System

of Systems Trend
The role of mechatronic systems in the system of systems (SoS) can be defined
as “large-scale integrated systems that are heterogeneous and independently oper-
able on their own, but are networked together for a common goal” [43]. Systems
participating in a SoS are independently designed and can operate autonomously
[44]. Following this definition, mechatronic systems and CPS can be considered as
components of SoS (see Chap. 10).
Main characteristics of SoS are defined in [44]. A SoS performs functions

which are not achievable by single systems (so called emergent behaviour). SoS
are geographically distributed, such that all participating systems need to exchange
data in a distributed and remote way. This implies the emphasis of previous sec-
tions on CPS and IoT trends and the next generation of connected mechatronic
systems. Another characteristic is the evolutionary development. This means that
the SoS offered services and purposes change over time and mechatronic systems
must be able to enter or leave the SoS during run-time. Managerial independence
is another characteristic of a SoS. The mechatronic systems are design indepen-
dently and manage their own goals.
This implies new methods and tools to develop SoS that could constrain
mechatronic systems design. SoS engineering deals with planning, analyzing,
organizing and integrating the capabilities of a mix of existing and new systems
into an SoS capability greater than the sum of the capabilities of the constituting
parts [45]. New mechatronics systems should enable this SoSE, with standardized
interfaces and cooperation enhancement.
6.3.4 The Role of Mechatronics in the Servitization Trend
In order to improve customer loyalty and to optimize the balance between the offer
and the customer’s requirements, significant numbers of industries are shifting from
a product centric approach to a bundle of products associated with and to services.
This is presented as a way to achieve objectives such as risk reduction, competi-
tiveness exposure reduction and sustainability through business model evolution.
The shift towards an integrated offer of products and services is illustrated by both
concepts of productization and servicization, in a paradigm of a transition from
a service or a product to an “integrated bundle of products and services” called a
product-service system (PSS) [46]. These evolutions are presented in Fig. 6.7.
Fig. 6.7 Evolution of the

PSS model [3]
In this context, mechatronics systems have a great role to transform prod-

uct into a platform supporting services. Even if the system is neither connected
all the time (IoT) nor able to interact autonomously with other systems (SoS),
the fact mechatronic systems is able to transform physical product into an intel-
ligent system allows to envision new perspectives. Some authors recently pre-
sented some approaches to consider the service during the mechatronic system
design [47].
6.3.5 The Role of Mechatronic Systems in the Factories

of the Future
Regarding the challenges of the factories of the future (FoF), the mechatronic sys-
tems can be viewed as the backbone for smart integration of the new factory mod-
els which is also strongly supported by the vision of Industry 4.0 [48, 49].
Since the end of 2000, the intensive use of digital factory technology [50]
allows the design of agile production lines and complex manufacturing systems
fully integrating mechatronics systems based on smart sensors, actuators, driv-
ers and controllers allowing the communication machine to machine, the remote
control of manufacturing operations, the self-diagnosis faults before failure of sys-
tems and an efficient management of energy usage in the production plants. In fac-
tory and production process environments, virtualization of operations thanks to
embedded mechatronic systems and large industrial internet connection via dis-
tributed networks and cloud computing enables the implementation and control of
cloud manufacturing operations and services [51–53].
Increasingly, new machine tools, industrial robots, and production equipment
are definitively based on mechatronic technology such as sensors, actuators d rivers
and controllers. Then, they become more and more autonomous to collect data
and information for monitoring of operations and remote controlling of processes.
Based on all these distributed manufacturing information and data, manufacturing
execution systems (MES) will work in real time to enable an efficient, agile and
flexible production management based on information and control alignment and
interoperability with the enterprise resources planning (ERP) system [54, 55].
Benefits will be obtained in improving productivity, supply chain management,
resource and material planning and product lifecycle management with the com-
plete integration of information and communication technology (ICT) and indus-
trial internet as support of enterprise information system [56, 57].
All these contributions based on the backbone of mechatronics integration in
the factory and manufacturing plants will ensure the future generation of cyber-
physical production systems and support the architecture of systems of manufac-
turing systems [58].
6.4 Conclusions
The chapter has dealt with new models for design processes of mechatronic sys-
tems and their future trends with novel applications. After a detailed presentation
of the current models and standards of development processes for mechatronic
systems engineering, the future trends for mechatronic system design are dis-
cussed. First, the new developments of mechatronics in the field of CPSs were
considered. Second, the added-value of mechatronic systems for the implementa-
tion of the IoT has been detailed. Third, the role of mechatronics in the design and
integration of SoS was presented as last research trend.
Last, regarding the applications, two very interested topics were considered
with the integration of mechatronic systems in the servitization of products, on the
one hand. On the other hand, the generalization of mechatronics as backbone of
the FoF was discussed.
The future trends and models for the design processes of mechatronic systems
have to be considered as unquestionable enablers for transformation of complex
systems into CPSs or the global integration of the IoT. These design processes for
mechatronic engineering have to support the development of the new services or
the implementation of industrial internet for the FoF.
References
1. Eigner M (2014) Überblick Disziplin-spezifische und -übergreifende Vorgehensmodelle,

Modellbasierte virtuelle Produktentwicklung. Springer, pp 15–52
2. Sommerville I (2010) Software engineering, 9th edn. Addison Wesley, New York
3. Baines TS, Lightfoot HW, Evans S, Neely A, Greenough R, Peppard J, Roy R, Shehab
E, Braganza A, Tiwari A, Alcock JR, Angus JP, Bastl M, Cousens A, Irving P, Johnson
M, Kingston J, Lockett H, Martinez V, Michele P, Tranfield D, Walton IM, Wilson H
(2007) State-of-the-art in product-service systems. Proc IMechE Part B J Eng Manuf
221:1543–1552
4. Fogliatto FS, da Silveira GJC, Borenstein D (2012) The mass customization decade: an
updated review of the literature. Intl J Prod Econ 138:14–25
5. Isermann R (2007) Mechatronic design approach. The mechatronics handbook. CRC Press,
Boca Raton, pp 2–3
6. Hyman BI (2003) Fundamentals of engineering design, 2nd edn. Prentice Hall, Upper Saddle
River
7. Beitz W, Pahl G, Wallace K (2003) Engineering design: a systematic approach. Springer, Berlin
8. Ullman DG (2010) The mechanical design process. McGraw-Hill Higher Education, New
York
9. Howard TJ, Culley SJ, Dekoninck E (2008) Describing the creative design process by the
integration of engineering design and cognitive psychology literature. Des Stud 29:160–180
10. Blanchard BS (2012) System engineering management. Wiley, London
11. OMG Group (2007) OMG systems modeling language. OMG SysMLTM
12. Plateaux R, Choley J-Y, Penas O, Riviere A (2009) Towards an integrated mechatronic
design process. IEEE Intl Conf Mechatron, 1–6
13. Graignic P, Vosgien T, Jankovic M, Tuloup V, Berquet J, Troussier N (2013) Complex sys-
tem simulation: proposition of a MBSE framework for design-analysis integration. Procedia
Comput Sci 16:59–68
14. Ye Y, Jankovic M, Bocquet J-C (2012) Managing semantic consistency in model-based sys-
tems engineering using a matrix structure. In: 32nd ASME computers and information in
engineering conference, parts A & B, vol 2, pp 1193–1204
15. Zheng C, Bricogne M, Le Duigou J, Eynard B (2014) Survey on mechatronic engineering: a
focus on design methods and product models. Adv Eng Informatics 28:241–257
16. Chen K, Bankston J, Panchal JH, Schaefer D (2009) A framework for integrated design of
mechatronic systems. In: Wang L, Nee AYC (eds) Collaborative design and planning for dig-
ital manufacturing. Springer London, pp 37–70
17. Hehenberger P (2014) Perspectives on hierarchical modeling in mechatronic design. Adv
Eng Informatics 28:188–197
18. Layoutsynthese elektronischer Schaltungen - Grundlegende Algorithmen für die
Entwurfsautomatisierung (2006). Springer, Berlin
19. Royce WW (1970) Managing the development of large software systems. In: Proceedings of
IEEE western electronic show and convention (WesCon), Los Angeles, pp 1–9
20. Boehm BW (1988) A spiral model of software development and enhancement. Computer
21:61–72
21. Forsberg K, Mooz H (1992) The relationship of systems engineering to the project cycle. Eng
Manag J 4:36–43
22. US Department of Transportation (2007) Systems engineering for intelligent transportation
systems, Washington
23. Aca J, Ramos M, Serrano JL, Ahuett H, Molina A (2006) Concurrent engineering of
mechatronic products in virtual enterprises: selection and deployment of a PLM system for
the machine tool industry. In: Luo Y (ed) Cooperative design, visualization, and engineering.
24. Lefèvre J, Charles S, Bosch-Mauchand M, Eynard B, Padiolleau E, Multidisciplinary model-
ling and simulation for mechatronic design. J Des Res 12(1–2):127–144
25. Alvarez Cabrera AA, Foeken MJ, Tekin OA, Woestenenk K, Erden MS, De Schutter B, van
Tooren MJL, Babuška R, van Houten FJAM, Tomiyama T (2010) Towards automation of
control software: a review of challenges in mechatronic design. Mechatronics 20:876–886
26. Bathelt J, Jonsson A, Bacs C, Dierssen S, Meier M (2005) Applying the new VDI design
guideline 2206 on mechatronic systems controlled by a PLC. In: International conference on
engineering design (ICED 05), Melbourne, p 2601
27. Mhenni F, Choley J-Y, Penas O, Plateaux R, Hammadi M (2014) A SysML-based methodol-
ogy for mechatronic systems architectural design. Adv Eng Informatics 28:218–231
28. Zheng C, Le Duigou J, Bricogne M, Hehenberger P, Eynard B (2015) Multidisciplinary inter-
face modelling: a case study on the design of 3d measurement system. In: 12th international
conference on product lifecycle management, Doha
29. Agile Manifesto (2015). www.agilemanifesto.org/. Accessed 20 Nov 2015
30. Cooke AG, Bonnema GM, Poelman WA (2012) Agile development for a multi-discipli-
nary bicycle stability test bench. In: 13th mechatronics forum international conference
(MECHATRONICS 2012), pp 812–819
31. Glas M, Ziemer S (2009) Challenges for agile development of large systems in the aviation
industry. In: Proceedings of 24th ACM SIGPLAN conference on companion on object ori-
ented programming systems languages and applications (OOPSLA’09), p 901
32. Stelzmann E (2012) Contextualizing agile systems engineering. IEEE Aerosp Electron Syst
Mag 27:17–22
33. Bricogne M, Rivest L, Troussier N, Eynard B (2014) Concurrent versioning principles
for collaboration: towards PLM for hardware and software data management. Int J Prod
Lifecycle Manag 7:17–37
34. Fisher J (1998) Model-based systems engineering: a new paradigm. Incose Insight, Incose
Insight 1:3–16
35. Object management group: systems modeling language specification. www.omg.org/
spec/SysML/1.2/PDF/. Accessed 20 Nov 2015
36. Cao Y, Liu Y, Paredis CJJ (2011) System-level model integration of design and simulation for
mechatronic systems based on SysML. Mechatronics 21:1063–1075
37. Kerzhner AA, Paredis CJJ (2012) A SysML-based language for modeling system-level archi-
tecture selection decisions. In: ASME 2012 conference on international design engineering
technical and computers and information in engineering, Chicago, pp 1263–1276
38. Rajkumar R, Lee I, Sha L, Stankovic J (2010) Cyber-physical systems: the next computing
revolution. In: Proceedings of 47th design automation conference (DAC’10), p 731
39. Rajkumar R (2012) A cyber-physical future. Proc IEEE 100:1309–1312
40. Kopetz H (2011) Real-time systems. Springer, Berlin
41. Bradley D, Russell D, Ferguson I, Isaacs J, MacLeod A, White R (2015) The internet of
things—the future or the end of mechatronics. Mechatronics 27:57–74
42. Vasi VS, Lazarevic MP (2008) Standard industrial guideline for mechatronic product design.
FME Trans 36:103–108
43. Jamshidi M (2008) Systems of systems engineering: principles and applications. Boca Raton,
USA
44. Maier MW (1996) Architecting principles for systems-of-systems. INCOSE Int Symp
6:565–573
45. Department of Defense (2004) Systems engineering, Defense Acquisition Guidebook, p 520
46. Mahut F, Daaboul J, Bricogne M, Eynard B (2015) Survey on product-service system appli-
cations in the automotive industry. IFAC-Pap Line 48:840–847
47. Pezzotta G, Pirola F, Pinto R, Akasaka F, Shimomura Y (2015) A service engineering frame-
work to design and assess an integrated product-service. Mechatronics 31:169–179
48. Zuehlke D (2010) Smart factory—towards a factory-of-things. Ann Rev Control 34:129–138
49. Eynard B, Bosch-Mauchand M (2015) Integrated design and smart manufacturing, concur-
rent engineering: research and applications. doi:10.1177/1063293X15607367
50. Chryssolouris G, Mavrikios D, Papakostas N, Mourtzis D, Michalos G, Georgoulias
K (2009) Digital manufacturing: history, perspectives, and outlook. J Eng Manuf
223(5):451–462
51. Tao F, Zhang L, Venkatesh VC, Luo Y, Cheng Y (2011) Cloud manufacturing: a computing
and service-oriented manufacturing model. J Eng Manuf 225(10):1969–1976
52. Wu D, Greer MJ, Rosen DW, Schaefer D (2013) Cloud manufacturing: strategic vision and
state-of-the-art. J Manuf Syst 32(4):564–579
53. Gao R, Wang L, Teti R, Dornfeld D, Kumara S, Mori M, Helu M (2015) Cloud-enabled
prognosis for manufacturing. CIRP Ann Manuf Technol 64(2):749–772
54. Belkadi F, Troussier N, Eynard B, Bonjour E (2010) Collaboration based on product lifecy-
cles interoperability for extended enterprise. Intl J Interact Des Manuf 4(3):169–179
55. Agostinho C, Ducq Y, Zacharewicz G, Sarraipa J, Lampathaki F, Poler R, Jardim-Goncalves
R (2015) Towards a sustainable interoperability in networked enterprise information systems:
trends of knowledge and model-driven technology, Computers in Industry. doi:10.1016/
j.compind.2015.07.001. Accessed 20 Nov 2015
56. Le Duigou J, Bernard A, Perry N (2011) Framework for product lifecycle management inte-
gration in small and medium enterprises networks. Comput Aided Des Appl 8(4):531–544
57. El Kadiri S, Grabot B, Thoben KD, Hribernik K, Emmanouilidis C, von Cieminski G,
Kiritsis D (2015) Current trends on ICT technologies for enterprise information systems.
Comput Ind. doi:10.1016/j.compind.2015.06.008
58. Lee J, Bagheri B, Kao HA (2015) A cyber-physical systems architecture for industry 4.0-
based manufacturing systems. Manuf Lett 3:18–23
Chapter 7
Design of Biomimetic Soft Underwater
Robots
Aiguo Ming and Wenjing Zhao
7.1 Introduction
The evolution process for creatures is very, very long and contains many useful
secrets and rationality mostly hidden in their structure, motion and configuration.
Biomimetic mechatronic design is a useful approach for future mechatronic inno-
vation, which can significantly enhance the performance of mechatronic systems.
As an important issue for biomimetic mechatronic design, it is necessary to make
a robot as soft as a natural creature to achieve more efficient, high-performance
and creature-like motions. Compared to a conventional rigid robot, the design and
control of a soft robot is difficult because the coupling between the flexible struc-
ture and surrounding environment should be considered, which is very difficult to
resolve due to the large deformations and complicated and coupled dynamics. This
is the main reason why design methods for soft robots have not been established,
despite the many trial developments of soft robots that have been undertaken to
date. The challenges for the design of biomimetic soft underwater robots based on
numerical simulation considering the coupling between the flexible structure and
surrounding fluid as well as control technologies are described in this chapter.
7.2 Basic Theory and Developments for Underwater

Robots
Humans have had dreams for robots for many years, hoping that people can be
replaced by robots for some work or tasks. In 1954, the first robot with point-to-point
control was implemented by George Devol in the United States [1]. From then, the
A. Ming (*)
The University of Electro-Communications, Tokyo, Japan
W. Zhao
Zhejiang University of Technology, Hangzhou, China

DOI 10.1007/978-3-319-32156-1_7
92 A. Ming and W. Zhao
robot has come into human life and people’s dreams begin to come true as robots
start to play more and more important roles in many fields.
Due to the unknown underwater environment, many underwater robots have
been designed and developed for seabed exploration, observation, leakage detec-
tion, military reconnaissance, rescue, environment protection, etc. [2–5]. Because
of the complicated underwater environment and tasks, underwater robots need to
have better mobility and motion performance in their special working environ-
ment. Underwater robots with simple moving mechanisms such as wheels and
crawler tracks have not been able to meet the demands of the various complicated
environments and tasks. These kinds of robots also have poor adaptability and
flexibility, which can cause enormous energy deficiencies in their propulsion.
Today, many researchers are looking for biological inspiration to create a new
generation of swimming automatons known as biomimetic underwater robots to
achieve high efficiency, good mobility and manoeuvrability. A biomimetic under-
water robot has to mimic the external shape, body structure, motion principles and
behaviours of creatures in the underwater environment, and is able to undertake
the work based on the creature’s motion characteristics [5]. Engineers and scien-
tists have thus concentrated on the design and development of biomimetic under-
water robots mimicking swimming creatures [6–25].
As one of the most active fields in the development of science and technol-
ogy, biomimetic underwater robots are achievements of interdisciplinary science
including robot technology, microfabrication technology, biological science and
hydrodynamics. Essentially, the biomimetic underwater robots are complicated
systems ranging from metre to nanometre scale [10]. Many inorganic components
(such as mechanical, electrical, hydraulic components) and organic bodies are
utilized to build the robot system. The design of biomimetic underwater robots is
particularly difficult for the operation and integration of these components to work
efficiently, reliably and autonomously in the underwater environment.
The research on biomimetic underwater robots mainly encompasses two areas:
the study of the basic theory and the development of biomimetic underwater
robots with high speed, efficiency and mobility.
7.2.1 Propulsion Theories
In nature, there are thousands of species and forms of aquatic organisms. They have
various morphologies and motion patterns and can be classified into three types:
first, cilia propulsion used by many protozoa and coelenterates on the body surface;
second, jet propulsion as with jellyfish and squid using the recoil force for propul-
sion by jetting fluid in the opposite direction; third, undulation propulsion where
aquatic organisms swing their body to achieve the propulsive force. A large num-
ber of fish use this undulation approach for propulsion by swinging their body or
caudal fins, pectoral fins and so on. This undulation propulsion is the main focus in
much research into designing and developing biomimetic underwater robots.
7 Design of Biomimetic Soft Underwater Robots 93
Undulation propulsion is different from the propulsion mechanism of a

conventional propeller. Reynolds number is a dimensionless number describing
the relative importance between forces from fluid viscosity and fluid inertia, and
is generally described by Eq. 7.1. According to the different values of Reynolds
number, fluid flows are divided into different patterns. When Re < 1, the fluid vis-
cous force dominates, and if Re is larger than 1000, the fluid inertial force domi-
nates in fluid flow [26].
ρUL
Re = (7.1)
µ
where U is the forward swimming speed of the fish; L is the length of the fish; ρ is
fluid density and μ is fluid dynamic viscosity.
The swimming kinematics of underwater creatures involves different propul-
sion mechanisms, such as meandering propulsion with a travelling wave. These
kinematics result from the complicated coupling dynamics between the forces
from muscles and from the surrounding fluid. In the swimming motion of a real
underwater creature, the muscles deform the soft body, which then applies a force
to the surrounding fluid. In return, the fluid also applies a force on the soft body
which changes the body shape. This interaction is repeated endlessly and propul-
sive force is generated as its result.
In order to model the deformation of a soft body from real creatures accurately,
the forces mentioned above must be taken into account simultaneously. Few mod-
els have been proposed to solve this coupling problem due to its complexity [27].
Instead, according to the different swimming motions of real creatures, most theo-
ries define the creature body as a given shape to predict the fluid forces, such as
theories used in fish-like robot design with the basic propulsion model classified
either as a resistive model or a reactive model.
The resistive model is proposed by Taylor [28]. The thrust is estimated from
the sum of the drag forces on the fish body with a travelling wave whose speed is
larger than the forward swimming velocity. The body is divided into many seg-
ments along the body length. The drag force of each segment can be calculated by
Eq. 7.2:
D = 0.5Cd ρu2 s (7.2)
where Cd is the drag coefficient; ρ is fluid density; s is surface area of segment;
u is swimming velocity including contributions from forward motion and lateral
undulation.
The resistive model is built based on a quasistatic method using steady flow
theory to predict the fluid force in the swimming motion. In this model, because
the fluid inertial force is neglected, its applicability is restricted to a low Reynolds
number. And because it is assumed that anterior segments of the creature do not
affect fluid flow moving past the body, this model may not hold for the undulation
of real fish. For the anguilliform type, because viscous forces play significant role,
it belongs to this resistive model. The problems of the resistive models are dis-
cussed in detail by Webb [29]. Although this resistive model has some limitations
for estimating resistive forces, it is important in the thrust generation of real crea-
tures whose drag force plays the dominant part in the interaction with surrounding
fluid [30] and may reduce the propulsive efficiency [29].
Later, the reactive model is presented to design a biomimetic underwater robot
with a fish-like propulsion motion. The model deals with more realistic fish-type
motions, assuming an inviscid fluid, that is, the fluid viscosity is assumed to be
negligible. A two-dimensional waving plate theory was developed originally by
Wu [31]. Based on the slender body theory stemming from aerodynamics, the
elongated-body theory is formed and it is suitable for subcarangiform and carangi-
form propulsion modes of real fish.
The favourite mathematical model of fish swimming used in the design of a
biomimetic underwater robot is the elongated-body theory developed by Lighthill
[32, 33]. In this model, the force from the fluid induced by the lateral acceleration
and deceleration of the undulating body is estimated based on acceleration reac-
tion in which the surrounding fluid is regarded as added mass [34, 35]. Thus, this
model is generally termed as the reactive model. Ignoring fluid viscosity, the aver-
age thrust power can be calculated using Eqs. 7.3–7.6 [32, 36]. The average thrust
force can be approximated by thrust power divided by forward swimming speed.
PT = mwUW − 0.5mw2 U (7.3)
2
d
m= πρ (7.4)
2
U
w=W −W (7.5)
V
πfA
W= (7.6)
1.414
where PT is the thrust power; m is the added virtual mass per unit length; w is the
velocity of water at the trailing edge; W is the rms value of the lateral velocity of
the trailing edge; U is the forward swimming speed; V is the velocity of the body
propulsive wave; ρ is the density of water; d is the span of the caudal fin at the
trailing edge; f is the oscillating frequency of the caudal fin; A is the oscillating
amplitude of the caudal fin.
The reactive model described above assumes that the undulating amplitude is
small, so that the angles of the body with the swimming direction are close to zero.
When the angles become large, more energy is lost into the wake around the fish
body. A large-amplitude elongated-body theory developed by Lighthill considers
this effect and motion with arbitrary amplitude [5, 33]. This theory is better suited
to carangiform type of fish swimming, where the lateral motion amplitude of
the caudal fin is large. The elongated-body theory assumes that the inertial force
completely dominates in the swimming. It is not suitable for anguilliform type
where viscous forces play the key role. Besides, it cannot be applied to the uniform
mode because the shapes of caudal fin and pectoral fin violate the fundamental
assumption of slenderness. Nevertheless, elongated-body theory has been used
regularly in swimming studies and in robot design for predicting propulsive force
and efficiency [29, 37, 38].
7.2.2 Development
MIT developed the world’s first biomimetic robotic tuna (Robotuna) in 1994.
From then on, development of biomimetic underwater robots became a popular
topic with the combination and advancement of biomimetics, mechanics, mate-
rials and driving systems, and many fish robots have been developed. For exam-
ple, RoboPike and vorticity control unmanned undersea vehicle (VCUUV) from
MIT as the improved Robotuna [24, 39, 40]; robotic fish inspired by the carp
from Mitsubishi [41]; a robotic black bass with pectoral fin propulsion from Kato
Laboratory at Tokai University [42]; the robotic fish UPF-2001 by the National
Maritime Research Institute (NMRI) [43]; underwater robot with rigid tail from
Michigan State University; Nanyang Awana NAF-I and RoMan-II using caudal
fin and pectoral fin propulsion, respectively [16, 44]; snake-like robot AmphiBot
by the École polytechnique fédérale de Lausanne [18]; snake-like robot HELIX-I
of TIT; SSSA lamprey-like robot; underwater robot CUTTLEFIN with fin-based
undulating propulsion by Autonomous Systems Laboratory at ETH Zurich [45];
FILOSE robot from Tallinn University; autonomous robotic fish in the London
Aquarium from the University of Essex [19, 46]; robotic eel from the Methran
Mojarrad group are exemplars [5, 15, 17, 20].
These biomimetic underwater robots usually use real fish as biomimetic objects
with high performance. They are available and have various potential applications
in the aspects of protecting the underwater resources from pollution, underwater
rescue, observation of the seabed and other special underwater tasks. Among the
above-mentioned biomimetic underwater robots, the propulsion methodologies
are mainly by conventional motor mechanisms. The motor mechanism is simple,
but lacks flexibility. The robots using motor mechanisms usually are of large size
and heavy structure with rigid materials and complex control system [47, 48]. The
propulsion is not smooth as the real creatures, and results in low swimming effi-
ciency and flexibility [49, 50]. It is difficult to realize creature-like propulsion per-
formance with good flexibility by this kind of robots.
7.3 Biomimetic Soft Underwater Robots and System

Design Architecture
So far, the biomimetic approach has been used to design a soft underwater robot
for operation in the unknown or complicated environments. Through mimicking
a real creature, the functions of the soft underwater robots can be classified into
Fig. 7.1 Decomposition of functions for a biomimetic underwater robot
two types: moving like a real creature and looking like a real creature, as described
in Fig. 7.1. For movement like a real creature, propulsion modes and swimming
motions are focused on. In creature-like swimming, the performance criteria such
as velocity, swimming number (the distance moved by per tail beat, related to
velocity, frequency and body length), energy consumption, flexibility, adaptabil-
ity, payload, are adopted to evaluate the motion performance. For looking like a
real creature, matching body shape, colour patterns, etc., is implied. Colour pat-
terns are composed of pigment-based and structural colour patches. It can signify
the status of a creature or its motivation and can be used to communicate among
species.
For the design of soft underwater robots by mimicking the real creature, an
architecture is shown in Fig. 7.2. Frame members, body structure and materials,
actuation mechanism, control system and buoyancy modulation, need to be con-
sidered in the robot design. The classification of propulsion modes for the bio-
mimetic soft underwater robots is traditionally similar to that of real fish [5, 47,
51–56]. Based on the selected biomimetic creature’s propulsion mode and cor-
responding swimming motion described in Fig. 7.2a [5, 51, 53, 56], the robot
body structure is designed, with considering important factors such as robot’s
rigidity(softness), strength, stability, safety, payload, durability and so on. The
suitable soft materials also need to be adopted in robot body design.
As for the actuation of the soft underwater robot, the appropriate actuation
mechanism, such as by motors or flexible actuators with DC/AC source, should be
adopted for desired robot performance. Based on the actuation mechanism of the
soft underwater robot, the corresponding control systems are designed and con-
structed for its propulsion, including the communication system and driving sys-
tem. In order to realize the dive motion of the soft robot, the buoyancy modulation
needs to be investigated. The performance for orientation and depth of the dive
motion for the soft robot should be evaluated. According to the various demands
for soft biomimetic underwater robots in the underwater environment, different
Fig. 7.2 Architecture for the design of a biomimetic soft underwater robot. a Propulsion modes.
b System design architecture
components mentioned above can be highlighted to design the different soft bio-
mimetic underwater robots for different functions.
In the design of a high-performance biomimetic soft underwater robot, biology,
mechanics, hydrodynamics, control system and power source should be consid-
ered synergistically. In the mechanical domain, it is necessary to pay more atten-
tion to the robot frame, mechanical structure, soft materials and corresponding
actuation components using biomimetic approach.
In the underwater environment, fluid dynamics plays the key role in robot pro-
pulsion and it is very important to investigate the hydrodynamic performance
around the soft robot. It is necessary to utilize computational fluid dynamics
(CFD) and flow visualization in the robot design as well as hydrodynamic perfor-
mance evaluation. The vortex distribution in the wake around the robot has a great
impact on robot propulsion and the wake dynamics must be considered in order to
improve the propulsion mechanism to achieve a high performance.
Motion control and communication approaches are also necessary to achieve an
optimal control system for the designed robot. High-performance sensors are also
needed for feedback control and underwater navigation. For the power source, a
power source with high energy density, high efficiency, high reliability and high
safety is needed. In the long term, energy harvesting from the wave and the motion
of the robot is necessary.
Generally, it is difficult to reproduce a creature-like motion using the conven-
tional rigid mechanisms, and it results in low efficiency and mobility for propul-
sion [48, 57]. To solve the problem, it is very important to make the biomimetic
underwater robots as soft as real creatures, including using soft actuators. Many
biomimetic underwater robots use flexible actuators with smart materials for
robot actuation, to obtain relatively better propulsion characteristics and motion
performances similar to those of real creatures [50, 58–65]. In the early stage of
the design of a biomimetic soft underwater robot using flexible actuators, most
of adopted flexible actuators are ionic polymer–metal composite (IPMC), shape
memory alloys (SMA), electrostatic film, PZT film or piezoelectric fibre compos-
ite (PFC). A comparison of characteristics of the flexible actuators is shown in
Table 7.1. Among actuators, PFC actuator has larger output and faster response.
And the propulsion mechanism using PFC actuator can be constructed to a simple
and compact structure with high energy conversion efficiency and fast response
based on high voltage.
However, in most research using such flexible actuators, only the feasibility of
applying the flexible actuators to the biomimetic underwater robots to realize basic
creature-like swimming motions have been considered. That is, how to achieve
high motion performances of biomimetic soft underwater robots by design and
control based on the essence of biomimetic soft robots is not emphasized in these
works. A biomimetic soft underwater robot is intended to be soft and how to real-
ize the efficient interaction between the flexible structure and the fluid inspired by
biology is the essence of the biomimetic soft underwater robots.
Table 7.1 Performance comparison of typical flexible actuators

Item Response Durability Environment Output
IPMC Fast Good Water is essential Low
SMA Slow Good Cooling and temperature control large
Electrostatic film Slow Good High voltage Large
PFC Fast Good High voltage Large
Because biomimetic soft underwater robotics is a relatively young research

field, new problems for different components and whole system are posed compar-
ing with conventional theories and technologies of rigid robots in the underwater
environment. Equivalent theories and technologies for biomimetic soft underwa-
ter robots have not yet been defined in a general form and researchers are still
exploring new ways to develop the robots with high performance by trial-and-error
method. In particular, the design and control methods for both mobility, stability
and robustness with considering bioinspired compliance remain to be established.
The requirements for shape, driving, controllability, path planning and sensing are
complex and difficult in comparison with those for a rigid robot.
7.4 Proposal for a Numerical Simulation-Based System

for Developing Biomimetic Soft Underwater Robots
7.4.1 System Configuration and Scheme
Development of biomimetic soft underwater robots is currently being carried out

basically by trial and error and results in high cost for both robot development and
performance improvement [66].
Some underwater robots have already benefited from design using available
simulators [67, 68]. However, in the previous work on underwater robot develop-
ment, robot design is carried out almost without utilizing the interaction between
the flexible structure of robot and the surrounding fluid to realize propulsion
performance by simulation [66]. Additionally, the control technologies of the
designed robots are considered separately with the design of the robot [10].
In order to solve such problems, a design method different from that of pre-
vious work is proposed. It emphasizes the interaction between the flexible robot
structure and the surrounding fluid based on a numerical simulation method and
integration of the design problem with control problem.
Numerical simulation can consider the actual robot model including the
mechanical structure with large deflection and the interaction with surrounding
fluid effectively for performance estimation. Based on considering the interac-
tion between flexible structure and surrounding fluid as well as the control input,
the creature-like propulsion motion can be reliably established, and whether the
design requirement is satisfied or not can be verified by the simulation results.
The hydrodynamic performance and propulsion characteristics also can be identi-
fied and evaluated in the numerical simulation system by comparison with results
of the experiment. The feasibility of the robot design can be carefully estimated
through the numerical modelling and simulation analysis for final behaviour
design and applications. When the desired propulsion characteristics and hydrody-
namic performances are identified, the designed robot will be adopted as the suit-
able robot model for prototype fabrication.
Fig. 7.3 Numerical simulation-based system for developing biomimetic soft underwater robot
A numerical simulation-based system, in which the interaction between the

flexible robot structure and fluid as well as control input are taken into account, is
hereby proposed as a necessary prerequisite for the successful design and develop-
ment of the desired biomimetic soft underwater robot. Figure 7.3 shows the con-
struction of the numerical simulation-based system for robot development. In the
system, the optimal robot model is firstly to be derived through numerical simula-
tion and analysis. Then the fabrication and experimental evaluation are carried out.
In order to develop a high-performance soft underwater robot using the biomi-
metic method, the creature-like propulsion approach must be determined based
on different propulsion modes of real creatures. Through the designed propul-
sion approach, the modelling of the robot including the flexible structure, driving
model and surrounding fluid is performed. Based on this robot model and control
input design, the numerical coupling simulation considering the large deflection
of structure and its interaction with surrounding fluid is performed to realize and
establish the creature-like propulsion motion of designed robot, and then robot
structure is optimized by established coupling simulation for performance
improvement until achievement of the optimal model with desired performances.
Through the optimization using coupling simulations, it can effectively obtain a
relatively optimal model. When establishing the optimal model, experimental
evaluation will be done by using an actual prototype. The detailed scheme of the
design of a biomimetic soft underwater robot using the numerical simulation sys-
tem is set out in Fig. 7.4.
Through design requirements of desired robot, an appropriate real creature is
selected and referred to as a biomimetic object. According to geometric character-
istics and swimming features of chosen real creature, the morphological and kine-
matic parameters are extracted to build the numerical model of the robot. Then, in
terms of the requirements of hydrodynamic performance and manoeuvrability, the
structural morphological parameters such as structural shape, stiffness and mass
distribution, are determined initially through the numerical coupling simulation to
Fig. 7.4 Detailed schema for the design of a biomimetic soft underwater robot
derive the suitable kinematic parameters. Based on this initial robot model, modi-
fying the parameters conditioned by both structure and fluid domain to design
again for establishment of the structural constrains, control input and fluid bounda-
ries until desired propulsion characteristics are realized.
Next, through the established boundary conditions in the structure and fluid
domains and feasible control input, the optimization using coupling simulation is
performed until an optimal model with high performance is achieved. Based on an
optimal robot model, prototype fabrication and experimental evaluation are finally
carried out.
It is possible to effectively develop the high-performance soft biomimetic
underwater robot using this detailed scheme for robot design in a numerical sim-
ulation system. By the numerical simulation, numerical models can be further
modified to improve the robustness and stability of motion due to environmental
uncertainty.
7.4.2 Case Study: Development of a Subcarangiform Robot
Based on the method for designing biomimetic soft underwater robots mentioned
above, a simple biomimetic soft underwater robot has been designed. Figure 7.5
shows the components of the robot. The fish type with BCF propulsion is regarded
as the biomimetic object, and the subcarangiform swimming mode is adopted.
In order to achieve the smooth propulsion and good flexibility, the flexible MFC
actuator (a typical PFC actuator) is utilized to design the actuation structure of the
soft robot. The corresponding driving system is also built to meet the requirements
Fig. 7.5 Components of biomimetic soft underwater robot
of the flexible MFC actuator, including the control input design of basic signals.
Based on the characteristics of the flexible MFC actuator, the carbon fibre-rein-
forced polymer (CFRP) is used as the main material for the robot body to realize
desired deformation of the robot. Different with the conventional design method
in which only the body structure is concerned, the fluid effect and interaction
between robot structure and fluid are considered together with control input in
simulation for design.
Using the proposed system and detained scheme shown in Figs. 7.3 and 7.4,
respectively, the biomimetic soft robot has been designed and developed [69].
Figure 7.6 then shows the prototype of the designed biomimetic soft underwater
robot and it will be called as new robot hereafter. A CFRP plate sandwiched
by two MFC plates (M-8528-P1 type) is used as actuator structure for bending
deformation. A weight made of steel is placed on robot head to increase the
displacement of the tail end. A float is placed on the top part of prototype for
balancing the robot in the liquid. The detailed specifications of the prototype are
shown in Table 7.2. The epoxy 3M-DP460 is used to bond the MFC plate onto
CFRP plate. The robot is put in a cubical fluid tank (590 × 133 × 440 mm) filled
by Fluorinert Electronic Liquid FC-3283 with considering the high voltage driving
of the MFC. For comparison, a robot developed in [70] is adopted and it will be
called as old robot hereafter.
In the experiment a high-speed camera is used for observation and meas-
urement of the robot motion. The driving system of the prototype is shown in
Fig. 7.7. The control signal is generated by the computer and basic signal wave-
forms such as sine, square and triangle are used. A voltage follower between the
high-voltage amplifier (AMP PA05039) and the computer is used to match the
impedances. The high-voltage amplifier outputs −500 V to +1500 V to MFCs
according to the control reference from the computer through DA board.
As the results of experiment, the new robot can realize various swimming
motions in the fluid, such as forward motion, backward motion, turning motion,
etc. The swimming velocities of the new robot and the old robot at different
frequencies ranged from 1 to 30 Hz are shown in Fig. 7.8. The input voltage in
14mm 64mm 3mm 60mm
Fig. 7.6 Prototype of new robot
Table 7.2 Specifications of Item Specification

new robot
Caudal peduncle height (mm) 3
Caudal fin height (mm) 60
Radius of head weight (mm) 7
Maximum body height (mm) 64
Type of actuator MFC 8528P1 × 2
Actuator dimensions (mm) 112 × 40
Actuator active area (mm) 85 × 28
Body thickness (mm) CFRP-0.2
Adhesive Epoxy 3M-DP460
square waveform with the range of −500 V to +1500 V is applied on both actua-
tors. From the figure, it can be known that the maximum swimming velocity of the
new robot is 0.792 m/s at 25 Hz and it is larger than that of old robot (0.72 m/s).
In the biomimetic field, swimming number Sw of the fish is widely used to
evaluate the swimming performances. Sw can be expressed by Eq. 7.7, where V
is velocity, f is frequency, L is body length. Swimming number describes the dis-
tance of fish moved by per tail beat. The swimming number of the fishes is gener-
ally about 0.6 for high performances with good flexibility and mobility [71]. It can
be utilized to evaluate the fish-like swimming performance of the biomimetic soft
robots. The calculated swimming numbers of the new robot and the old robot at
different frequencies are shown in Fig. 7.9.
Fig. 7.7 Driving system

of robot prototype in
experiment. a Platform.
b Driving system
Fig. 7.8 Comparison of 1.0

swimming velocities between Old Robot
New Robot
new robot and old robot at
different frequencies 0.8
Velocity (m/s)
0.6
0.4
0.2
0.0
0 5 10 15 20 25 30
Frequency (Hz)
Fig. 7.9 Comparison of 0.7

swimming numbers between Old Robot
new robot and old robot at 0.6 New Robot
different frequencies
0.5
0.4
Sw
0.3
0.2
0.1
0.0
0 5 10 15 20 25 30
Frequency (Hz)
V
Sw = (7.7)
fL
The swimming number of the new robot is larger than that of the old robot near
the resonant frequencies where the first three bending propulsion modes occur
(4, 21 and 29 Hz, respectively). Near the frequency of the first bending mode, the
swimming number is 1.1 times that of the old robot. And it is of the order of 1.8
times that of the old robot near the frequencies of the second and third bending
modes. The swimming number of the new robot is much closer to the value for a
real fish (about 0.6) when compared with that of the old robot. Thus, it can be con-
sidered that the new robot shows the better swimming performance.
Turning motion of the new robot is realized by applying the input voltage
signals with bias to the actuators of the robot. Figure 7.10 shows input voltage sig-
nals to actuators (both left and right actuators on robot body). The input voltage
of a square waveform is used to achieve a larger turning velocity than that by sine
waveform.
(a) Left (b) Left

Right
Right
1500 1500
Voltage (V)
1000 1000
Voltage (V)
500 500
0 0
-500 -500
0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.4 0.5
Time (s) Time (s)
Fig. 7.10 Input voltage on both MFC actuators for turning motion. a Left turn motion. b Right
turn motion
As a summary of the case study, a biomimetic soft underwater robot with

higher swimming velocity and swimming number similar to that of real fish has
been developed by optimization according to the design approach using numeri-
cal simulation system. It can be said that the design and control method by using
numerical coupling simulation analysis is useful and can be applied to evaluate
the propulsion performance of biomimetic soft underwater robots for further high-
performance robot development.
7.5 Conclusions and Future Challenges
The design of biomimetic soft underwater robots is discussed in this chapter.

Because most biomimetic underwater robots should adopt the flexible actua-
tors rather than motor mechanism, the importance of designing the biomimetic
soft underwater robots with high mobility, high efficiency and high adaptability
is emphasized. Since the design and development of biomimetic soft underwater
robots is a young field, theories and robotic technologies have not yet been refined
into a general method and robot researchers are still exploring new means of
achieving a new generation of biomimetic soft underwater robots.
The system design architecture for biomimetic soft underwater robot is pre-
sented, in which the biology, robotics, hydrodynamics, control system and power
source are the main components to be integrated. In order to develop a high-per-
formance biomimetic soft underwater robot based on design requirements and
associated functions, a developing system including numerical simulation consid-
ering the interaction between structure and fluid as well as control input design is
proposed as a necessary prerequisite for the design of a biomimetic soft underwa-
ter robot, because motion based on the efficient interaction between the flexible
structure and surrounding fluid by adequate control input is the essence for biomi-
metic soft underwater robots.
Using the detailed design scheme in the numerical simulation system, the struc-
tural constrains, control input and fluid boundaries can be established and an optimal
robot model can be achieved by simulation-based optimization. The environment
uncertainty, different fluid properties, patterns of fluid flow and corresponding
details should be considered in the numerical simulation system to improve robust-
ness and stability. A reliable, stable and accurate numerical simulation system can
accelerate the design process while realizing high propulsion performance.
However, the work up to now can be regarded just as a start point for designing
biomimetic soft robots with high performance. Towards realization of full biomi-
metic soft robots with advanced performance as creatures or beyond creatures, the
main challenges can be considered as follows.
1. From the view of biomimetics and biological evolution, biological structures
such as brain structure, skeleton distribution and nervous system play key roles
in realization of the behavioural features of real creatures. How to understand
and achieve the cognition of such behavioural features and their evolution from
the biological structure of real creature is a new direction and challenge in

the design of actuation mechanisms, control, navigation, communication and
modulation through the neurology and cognitive science for the development of
high-performance biomimetic soft robots.
Due to the great complexity of the biological musculoskeletal system of real
creatures, the achievement and utilization of biological musculoskeletal
distribution from a real creature through a biomimetic approach for desired
creature-like motion have not as yet been realized. The related theories
and design technologies have not been established in a general form. Future
applications of soft robots significantly need a focus on adaptive behaviour
design based on biomimetic approach during the motion design and control for
unknown and unstructured environments. Also, there remains the question as to
how to develop powerful and efficient artificial muscles and sensitive artificial
sensors with high S/N ratio, another unavoidable issue in realizing biomimetic
soft robots with high performance.
2. As the model for biomimetic soft robots, more complicated properties such
as nonlinear characteristics, non-holonomic constraints, continuum–discrete
conversion, etc., should be considered, because such properties are the
dominant factors governing the motions of creatures and robots. Besides the
important properties, the parameters for material property, soft actuators, etc.,
should be identified accurately because the parametric errors influence the
motion of the robot in a more direct way when compared to the rigid robot.
Further, although theoretically soft robots have infinite degrees of freedom, the
number of actuators and sensors in practical soft robot is finite. Many degrees
of freedom are not directly controllable.
How to locate the soft actuators and sensors in the structure is another impor-
tant issue in realizing controllable and efficient motions for the biomimetic soft
underwater robots.
3. As mentioned above, a biomimetic soft underwater robot is a continuum
with infinite elements and distributed actuators and sensors. And the robot is
located in water, analysis based on simulation with considering the fluid and
flexible structure interaction must be done for design and control of the robot.
Numerical simulation is a useful approach for the simulation. But, how to real-
ize efficient and stable motion analysis specially FSI analysis is a challenging
issue for the future, because it is difficult to get stable solution while with huge
computation cost by available software tools. Further, optimal design and opti-
mal control based on the numerical simulations should be performed.
4. A new manufacturing technology for biomimetic soft underwater robots should
be considered when designing the robots. Basically, the manufacturing tech-
nology for biomimetic soft robots is different from that of conventional rigid
robots. For example, 3-D printing technology is a potential technology for
fabricating biomimetic soft underwater robots, by embedding body materials,
actuators, sensors, etc., together. To develop the biomimetic soft robots with
high performance, precise 3-D printing and mounting technology capable of
multiple materials as well as module parts is expected to be developed.
5. Another challenging issue is how to enhance the autonomy of biomimetic soft

underwater robots. First, to realize the stable motion of the robots, internal
sensors such as motion sensors are needed to define the state of the robots as
necessary for the feedback control. Second, because the environment for under-
water robots is more complicated and time-varying, such as the complex flow
of water, many unstructured obstacles, sensors for recognizing the environment
are also necessary to control the motion of the robot in that environment. Using
the internal and external information from the sensors, self-stabilization, self-
compensation, self-adaption and self-diagnosing similar to those by creatures
are expected to be realized in the future.
The development of biomimetic soft underwater robots is a synergy technology
of biomimetics, soft robotics, materials science, intelligent control, and sens-
ing and communication technology. The technology is not only useful for devel-
oping underwater robots with advanced performance, but also can contribute to
the breakthrough of general robotic technology. With the remarkable progress
in related fields, it is believed that the high-efficient biomimetic soft underwater
robots with high mobility, adaptability and autonomy can be realized in the near
future.
References
1. Ded SR, Deb S (2010) Robotics technology and flexible automation. Tata McGraw Hill
Education, Noida
2. Ura T, Takagawa S (1994) Underwater robots (in Japanese). Seizan-do, Tokyo
3. Nagai M (1998) Hydrodynamics learning from dolphin (in Japanese). Ohmsha Ltd., Tokyo
4. Hirata K (2000) Development of experimental fish robot. In: Proceedings of the 6th international
symposium marine engineering, pp 711–714
5. Sfakiotakis M, Lane DM, Davies JBC (1999) Review of fish swimming modes for aquatic
locomotion. J Oceanic Eng 24(2):237–252
6. Ma G (2001) The research of the biomimetic robots. Robots 23(5):463–466
7. Kim E, Youm Y (2004) Design and dynamic analysis of fish robot: PoTuna. In: International
conference robotics and automation, pp 4887–4892
8. Deng X, Avadhanula S (2005) Biomimetic micro underwater vehicle with oscillating fin
propulsion: system design and force measurement. In: Proceedings of the IEEE international
conference robotics and automation, pp 3312–3317
9. Kumph JM (2000) Manoeuvring of a robotic pike. MS thesis, Massachusetts Institute of
Technology, USA
10. Yu J, Wang L, Tan M (2005) A framework for biomimetic robot fish’s design and its
realization. In: American control conference, pp 1593–1598
11. Zeng N, Hang GR, Cao GH, Wang ZL (2006) Present state and tendency of bionic
underwater robot. Mech Eng 4:18–22
12. Cai Z, Zou X (2004) Research on environmental cognition theory and methodology for
mobile robots. Robots 26(1):87–91
13. Liu J, Cai Z, Tu C (2002) Survey on evolutionary robotics. Control Theory Appl
19(4):493–499
14. Freska C, Brauer W, Habel C, Wender KF (2003) Spatial cognition: routes and navigation,
human memory and learning, spatial representation and spatial learning (Lecture notes in
artificial intelligence), vol 2685. Springer, Heidelberg
15. Triantafyllou MS, Techet AH, Hover FS (2004) Review of experiment work in biomimetic
foils. J Oceanic Eng 29(3):585–593
16. Low KH, Chong CW, Zhou C (2010) Performance study of a fish robot propelled by a
flexible caudal fin. In: International conference robotics and automation, pp 90–95
17. Tan X, Carpenter M, Thon J, Alequin-Ramos F (2010) Analytical modeling and experimental
studies of robotic fish turning. In: International conference robotics and automation,
pp 102–108
18. Crespi A, Badertscher A, Guignard A, Ijspeert AJ (2005) AmphiBot I: an amphibious snake-
like robot. Robot Auton Syst 50:163–175
19. Hu H (2006) Biologically inspired design of autonomous robotic fish at Essex. In:
Proceedings IEEE SMC UK-RI chapter conference advances in cybernetic systems, pp 1–8
20. Mojarrad M, Shahinpoor M (1997) Biomimetic robotic propulsion using polymeric artificial
muscles. In: International conference robotics and automation, pp 2152–2157
21. Anderson JM, Kerrebrock PA (1997) The vorticity control unmanned undersea vehicle
(VCUUV)—an autonomous vehicle employing fish swimming propulsion and manoeuvring.
In: l0th International symposium unmanned untethered submersible technology, pp 189–195
22. Hirose S, Fukushima EF (2004) Snakes and strings: new robotic components for rescue
operations. Int J Robot Res 23:341–349
23. Triantafyllou MS, Triantafyllou GS (1995) An efficient swimming machine. Sci Am
272:64–70
24. Mojarrad M (2000) AUV biomimetic propulsion. In: Oceans conference record, pp 2141–2146
25. Yu J, Tan M, Wang S, Chen E (2004) Development of a biomimetic robotic fish and its
control algorithm. IEEE Trans Syst Man Cyber Part B 34(4):1798–1810
26. Faber TE (1995) Fluid dynamics for physicists. Cambridge University Press, Cambridge
27. Cortez R, Fauci L, Cowen N, Dillon R (2004) Simulation of swimming organisms: coupling
internal mechanics with external fluid dynamics. Comp Sci Eng 6:38–45
28. Taylor GI (1952) Analysis of the swimming of long and narrow animal. Proc Royal Soc
Lond A 214:158–183
29. Webb PW (1975) Hydrodynamics and energetics of fish propulsion. Bull Fish Res Bd Can
190:1–159
30. Tytell ED (2004) The hydrodynamics of eel swimming II. Effect of swimming speed. J Exp
Biol 207:3265–3279
31. Wu TY (1961) Swimming of a waving plate. J Fluid Mech 10:321–344
32. Lighthill MJ (1960) Note on the swimming of slender fish. J Fluid Mech 9:305–317
33. Lighthill MJ (1971) Large-amplitude elongated-body theory of fish locomotion. Proc R Soc
Lond (B) 179:125–138
34. Batchelor GK (1973) An introduction to fluid mechanics. Cambridge University Press,
Cambridge
35. Daniel TL (1984) Unsteady aspects of aquatic locomotion. Am Zool 24:121–134
36. Webb PW (1992) Is the high cost of body caudal fin undulatory swimming due to increased
friction drag or inertial recoil. J Exp Biol 162:157–166
37. Muller UK, Stamhuis EJ, Videler JJ (2002) Riding the waves: the role of the body wave in
undulatory fish swimming. Integr Comp Biol 42:981–987
38. Tytell ED, Lauder GV (2004) The hydrodynamics of eel swimming. I. Wake structure. J Exp
Biol 207:1825–1841
39. RoboPike @ www.robotic-fish.net. Accessed 4 Nov 2015
40. Anderson JM, Chhabra NK (2002) Manoeuvring and stability performance of a robotic tuna.
Integr Comp Biol 42:118–126
41. Menzel P, Aluisio FD (2002) Robo sapiens: evolution of a new species. Art & Licensing
International, Manchester, pp 80–81
42. Kato N, Inaba T (1997) Hovering performance of fish robot with apparatus of pectoral fin
motion. In: 10th international symposium unmanned untethered submersible technology,
pp 177–188
43. Prototype Fish Robot UPF-2001 @ www.nmri.go.jp/eng/khirata/fish/experiment/upf2001/
index_e.html. Accessed 4 Nov 2015
44. Zhou C, Low KH (2010) Better endurance and load capacity: an improved design of manta
ray robot (RoMan-II). J Bionic Eng 7:137–144
45. Peter B, Ratnaweera R, Fischer W, Pradalier C, Siegwart RY (2010) Design and evaluation of
a fin-based underwater propulsion system. In: International conference robotics and automa-
tion, pp 3751–3756
46. Daily Mail (2009) Schools of robofish to sniff out pollution in the Thames @ www.dailymail.
co.uk/sciencetech/article-1163125/Schools-robofish-sniff-pollution-Thames.html. Accessed 4
Nov 2015
47. Breder CM (1926) The locomotion of fishes. Zoologica 4:159–297
48. Trivedi D, Rahn CD, Kier WM, Walker ID (2008) Soft robotics: biological inspiration, state
of the art, and future research. Appl Bion Biomech 5(3):99–117
49. Low KH, Willy A (2006) Biomimetic motion planning of an undulating robotic fish fin. J Vib
Control 12(12):1337–1395
50. Williams RB, Inman DJ (2002) An overview of composite actuators with piezoceramic
fibers. In: 20th International modal analysis conference, pp 421–427
51. Price AC, Weadick CJ, Shim J, Rodd FH (2008) Pigments, patterns, and fish behaviour.
Zebrafish 5(4):297–307
52. Roper DT, Sharma S, Sutton R, Culverhouse P (2011) A review of developments towards
biologically inspired propulsion systems for autonomous underwater vehicles. J Eng Marit
Environ 225(2):77–96
53. Lindsey CC (1979) Fish physiology—form, function, and locomotory habits in fish.
Academic Press, Cambridge
54. Webb PW (1984) Form and function in fish swimming. Sci Am 251:72–82
55. Low KH (2011) Current and future trends of biologically inspired underwater vehicles. In:
Defense science research conference and expo, pp 1–8
56. Fish FE (2001) Swimming in fish. Wiley, Hoboken
57. Liu H, Tang Y, Zhu Q, Xie G (2014) Present research situations and future prospects on bio-
mimetic robot fish. Int J Smart Sens Intell Syst 7(2):458–480
58. Chu WS, Lee KT, Song SH, Han MW, Lee JY, Kim HS, Kim MS, Park YJ, Cho KJ, Ahn
SH (2012) Review of biomimetic underwater robots using smart actuators. Int J Precis Eng
Manufact 13(7):1281–1292
59. Cho KJ, Koh JS, Kim S, Chu WS, Hong Y, Ahn SH (2009) Review of manufacturing
processes for soft biomimetic robots. Int J Precis Eng Manuf 10(3):171–181
60. Bhandari B, Lee GY, Ahn SH (2012) A review on IPMC material and actuators and sensors:
fabrication, characterization and applications. Int J Precis Eng Manuf 13(1):141–163
61. Kim B, Lee S, Park JH, Park JO (2004) Inchworm-like microrobot for capsule endoscope. In:
International conference robotics and biomimetics, pp 458–463
62. Hunter I, Lafontaine S, Hollerbach J, Hunter P (1991) Fast reversible NiTi fibers for use in
microrobotics. In: Micro electro mechanical system, MEMS’91, pp 166–170
63. Brunetto P, Fortuna L, Graziani S, Strazzeri S (2008) A model of ionic polymer-metal com-
posite actuators in underwater operations. Smart Mater Struct 17(2):025029–025040
64. Jayender J, Patel R, Nikumb S, Ostojic M (2008) Modeling and control of shape memory
alloy actuators. IEEE Trans Control Syst Technol 16(2):279–287
65. Shepherd R, Ilievski F, Choi W, Morin SA, Stokes AA, Mazzeo AD, Chen X, Wang M,
Whitesides GM (2011) Multigait soft robot. Proc Nat Acad Sci 108(51):20400–20403
66. Brutzman DP (1992) From virtual world to reality: designing an autonomous underwater
robot. In: AAAI fall symposium applications of artificial intelligence to real-world autonomous
mobile robots, 92-02, vol 1–5
67. Pappas G, Shotts W, O’Brien M, Wyman W (1991) The DARPA/Navy unmanned undersea
vehicle program. Unmanned Syst 9(2):24–30
68. Brutzman DP, Kanayama Y, Zyda MJ (1992) Integrated simulation for rapid development of
autonomous underwater vehicles. In: IEEE Oceanic engineering society conference, AUV
92, pp 3–10
69. Zhao W, Ming A, Shimojo M, Inoue Y, Maekawa H (2014) Fluid-structure interaction

analysis of a soft robotic fish using piezoelectric fiber composite. J Robot Mechatronics
26(5):638–648
70. Shintake J, Ming A, Shimojo M (2010) Development of flexible underwater robots with
caudal fin propulsion. In: IEEE/RSJ international conference intelligent robots and systems
(IROS), pp 940–945
71. Tanaka I, Nagai M (1996) Hydrodynamic of resistance and propulsion-learn from the fast
swimming ability of aquatic animals, Ship & Ocean Foundation, Tokyo, pp 14–19
Chapter 8
Improving the Robustness of Mechatronic
Systems
Antonio Lanzotti and Stanislao Patalano
8.1 Introduction
A system engineering approach seems to be effective in the design of mechatronic

systems [1] and over time, dedicated rules have been provided for the application
of such approaches to mechatronic systems design [2, 3]. Nevertheless, even if the
approach has been continuously used over an extended period of time, there is a
need for a revised methodological framework aimed to improve robustness of new
mechatronic systems, with respect to both conceptual and operational vulnerabilities.
The lack of robustness at conceptual level is mainly due to a violation of design
principles and a loss of knowledge. This lack is significant, especially in the case of
new systems when the design choices, related to the architecture of the subsystems,
have to be accomplished.
From the early development phases of new mechatronic systems, simulations
can be performed to improve and optimise design [4, 5]. Sophisticated multi-phys-
ical models are, in fact, currently developed thanks to the availability of integrated
environments for requirements specification, functional analysis, logical analysis
and geometrical modelling [4]. The use of bond graphs to model the behaviour of
systems characterized by subsystems operating with different energy domains is
an effective way to represent mechatronics systems. Several contributions to this
have been provided in literature [6, 7] and the existence of such contributions sup-
ports the modelling of mechatronic systems by this means, in particular to verify
limits and constraints in their application.
The lack of robustness at operational level is mainly due to the underestimation
of noise factors that can affect the performance of the system during the whole life
cycle. Understanding the variation of the complex system is the critical point of
robustness improvement.
A. Lanzotti · S. Patalano (*)

University of Naples Federico II, Fraunhofer JL IDEAS, DII, Naples, Italy

DOI 10.1007/978-3-319-32156-1_8
114 A. Lanzotti and S. Patalano
A significant issue within simulation activities is sensitivity analysis. The topic

of modelling, simulation and analysis of mechatronic systems, oriented to sen-
sitivity analysis, has been addressed in literature. El Fahime et al. [8] have stud-
ied the effect of variance related to dimensional parameters on the performances
of a mechatronic system using a Monte Carlo simulation. Precup and Preitl [9]
have presented a stability analysis method dedicated to fuzzy control systems with
mechatronics applications. Further, the authors provide for the sensitivity analysis
of fuzzy control systems with respect to the parametric variations of the controlled
plant for a class of servo-systems used in mechatronics applications.
If we consider a pure mechanical system it is possible to locate different
sources of variation. In fact, from a mechanical perspective we could consider
dimensional variations (lengths and angles), geometric form and feature varia-
tions (position, roundness, angularity and so forth) and kinematic variations (small
adjustments between mating parts). Therefore, according to a general architecture
of a mechatronic system [10, 11] and the presence of multidomain and connected
subsystems, it is possible to locate further sources of variation, related, for exam-
ple, to sensors and intelligent computer control. In this context, the improvement
of the robustness of a mechatronics system is an imperative need and research
activities related to Robust Design (RD) in engineering [12–14] could make a con-
crete contribution to the design of mechatronic systems.
The aims of the chapter are twofold. The first is to highlight the issue of
improving robustness of mechatronic system by working at conceptual level and
operational level, respectively. The second deals with the adaption of RD meth-
ods to improve robustness of a selected mechatronic system, by working on opera-
tional level within an automotive case study.
8.2 The Methodological Approach
If we consider a mechatronic system we could refer to a general system architec-

ture composed of a mainly physical subsystem, a mainly sensor subsystem and a
mainly control subsystem [10]. In the following, by addressing the general archi-
tecture, the characteristics of two main levels aimed at tackling the lack of robust-
ness are briefly described.
8.2.1 Tackling the Lack of Robustness

at the Conceptual Level
Working with mechatronic systems architecture and the set of above-mentioned

subsystems, several conditions appear. The system considered as a whole, in
fact, could present coupled design parameters, redundant design parameters or,
8 Improving the Robustness of Mechatronic Systems 115
Fig. 8.1 Design matrix of a mechatronic system with coupled parameters (top); iteration of FRs
refinements and zigzagging along V-model (down)
independent parameters. The last condition is the ideal one as stated by Axiomatic
Design (AD) or by Axiomatic Product Development Lifecycle (APDL) [15,
16]. In such a condition, the Design Matrix (A) that represents the link between
Functional Requirements (FRs) and Design Parameters (DPs) is a diagonal matrix.
Otherwise, the other conditions state the non-independence of FRs (Fig. 8.1) and
they require innovation on system architecture. In particular, design methods,
as for example the TRIZ method [16–20], involving technical or physical con-
tradictions, could be used and then the system architecture innovation could be
accomplished.
The way to apply such methods is represented by the system engineering
V-Model (Fig. 8.1). In particular, iteration from requirement analysis to imple-
mentation, moving along the “decomposition and definition line”, could be used to
accomplish the “refinement of FRs” and the “zigzagging between FRs and DPs”,
that are typical steps within AD [15].
Fig. 8.2 P-diagram for the ith subsystem of a mechatronic system (left) and for the whole
system (right)
8.2.2 Tackling the Lack of Robustness

at the Operational Level
The robustness at operational level could be tackled by considering both for a

mechatronic system as well as for the ith subsystem of a mechatronic system, a
parameter diagram, i.e. p-diagram (see Fig. 8.2) and the related Transfer Function.
In particular, the Transfer Function expresses the relation between Control plus
Signal Factors and Performances under the effect of the Noise Factors, within the
ith subsystem as well as the whole mechatronic system.
The relationship between the main components of a p-diagram and the compo-
nents introduced by AD (FRs, DPs and A) is the following. DPs are translated into
Control and Signal Factors; FRs are taken into account as Performances while the
Design Matrix (A) is represented by the Transfer Function. Finally, Noise Factors
are introduced as a perturbation related to DPs or external factors due to existing
boundary conditions.
Working within mechatronic system design two main goals have to be accom-
plished. A preliminary one deals with the knowledge of the “transfer function” to
be used. The transfer function could be: explicitly known (for example, function of
real variables and differential equations), implicitly known (for example, continu-
ous or discrete event simulation and numerical codes) or experimentally evaluated
(for example, applying statistical approach or neural networks). An actual chal-
lenging field of research is how to define reduced model that can be used at the
early stage of the design process of a new system instead of complete models that
require highly intensive use of experimental resources [21].
A successive goal is the understanding of variations to accomplish the robust-
ness of mechatronic system. The well-known RD methodology [22–24] is the
starting point of the approach aimed to improve robustness at operational level.
RD is a statistics-based methodology for improving the quality of a system by
reducing the sensitivity of its performances to the sources of variation during its
whole life cycle. Following that methodological path, after the system design
phase that is discussed in the previous paragraph, the critical phase is the so-called
“Parameter Design” phase.
The parameter design phase is aimed at identifying the main DPs, and to pre-
dict and evaluate their optimal settings. In this context, the attention is paid to
improving system robustness, i.e. the system insensitivity to the variation due
to noise factors. During the experimental phases, the design team can discover
interaction effects never thought and consequently can increase his technical
knowledge. Even if no “discovery” is attained, the experimental phase at least con-
tributes to improve the knowledge of the design team. DPs are variables whose
values can be still modified in this phase without any increasing of assembling or
usage cost.
After that, the so-called “Tolerance Design” phase allows reducing the effect
of variation allocating tolerances to parts in order to reduce the variation of the
responses at the minimum cost.
The key steps of parameter design phase according to the RD methodology are
as follows:
1. Choice of experimental design strategy
Experiments must be accurately planned. An effective and technically viable
experimental arrangement can be a cross-array [23] in which the “inner array”
is defined by the design settings obtained combining control and signal factors
and the “outer array” is defined by the settings of the noise factors levels. Each
design setting of the inner array is experimented several times, as prescribed by
the outer array. This is not the most efficient way to combine design and noise
factors (see for example [25]), but it is the classical one and is the easiest for
introducing the topic to engineers [24].
2. Choice of DPs and levels
Design parameters depend on the system architecture and are defined during
the preliminary design phase. It is interesting to understand at which extent
they are singularly effective (main effects) on improving performance, but also
to discover if there are synergistic or anti-synergistic interactions among these
parameters.
The choice of design parameter levels for an experiment is mainly left to the
judgment of designers/engineers, who will consider technical and economical
constraints (in the sense that the adoption of a certain design parameter level
should not significantly increase the manufacturing or usage cost).
3. Identification of noise factors and definition of levels
As already mentioned, the final objective of RD is to make the system design
setting as insensitive as possible (i.e. robust) to the variation of internal and
external characteristics, as for example part variation in their real parameters
compared to nominal ones. So in general, the variation around the nominal
value of each component of the system is considered as a noise factor in RD
terminology.
For the definition of the outer array, more levels of the noise factor must be cho-
sen. Obviously the higher the number of levels for the noise factor, the higher
will be the representativeness of the population variation, but the higher will be
the number of experiments to be performed in the virtual lab. So a compromise
solution must be found. The most widespread choice for the definition of noise
factor levels in the outer array is that proposed by Taguchi (see e.g. [22]) in
which two or three levels are chosen. This problem is known as a discretization
problem of a continuous distribution and many techniques have been proposed
to improve the estimation of the “weighted” loss function [12, 26].
4. Choice of the performance indicators (responses)
The responses for analysis are identified starting from the FRs, have a known
target and are measurable. So for each case, the loss function can be chosen
from among the classical ones or can be carefully tailored on the basis of the
available knowledge. When more performance indicators have to be taken
simultaneously into account, the problem moves from the univariate loss func-
tion to the multivariate case (see e.g. [14, 27–30])
5. Analysis of the experimental results and definition of optimal settings of main
design parameters
Experimental results can be analyzed on the basis of statistical methods. It can
be a more or less standard analysis, depending on the adopted experimental
arrangement and the degree of mathematical depth to be assured. Analysis of
main effects and Pareto diagrams are useful and simple tools to evaluate the
design parameters during explorative experimental phases [31, 32].
6. If the final quality level is not satisfactory, it is necessary to skip to the toler-
ance design phase
The classical tolerance design is the last chance to fulfil the FRs in spite of
variation but the smaller tolerance specification values, the higher are the pro-
duction costs [33–35].
8.3 Model Reduction and Robustness Analysis

for a Simple Case Study Concerning the Automotive
Power Window System
The aim of the case study is to apply some steps of the methodological framework
previously outlined to highlight actual and future research trends:
1. Development of the complete model to simulate the system acting as implicit
knowledge of the transfer function;
2. Execution of an experimental phase for the identification of the reduced
model through RSM methodology to simplify the task of optimal setting
identification;
3. Generation and evaluation of optimal design choices by means of the reduced

model taking into account noise parameters;
4. Validation of the results obtained by the reduced model through an experimen-
tal phase using complete model to verify optimal settings in order to consoli-
date robustness improvement.
8.3.1 The Power Windows System and the Related FRs
Power window systems are electromechanical devices used to lift and lower the
car window. The main subsystems/parts of such power window systems are a
DC motor, equipped with a worm gear, a sliding mechanism and a window con-
strained to move along rails. The DC motor–worm gear assembly is usually named
a “gear-motor” while the sliding mechanism usually characterizes the architecture
of the power window. When the command button is moved upwards or down-
wards, a voltage with direct or reverse polarity is applied to motor to drive the DC
motor in the forward or reverse direction.
The motor drives a worm gear that operates Bowden cables1 by means of a pul-
ley. Then, the cable moves one or two supports and, consequently, lifts and lowers
the automotive window. Usually, automotive power systems are single or double
Bowden type when the cable shapes a single loop or double loop (cross cable),
respectively. The present chapter deals with double Bowden power window
systems.
The FRs of the power window system that have to be assured, during the usual
development process of a new car, are the following:
FR1—the window must complete the stroke in a fixed time;
FR2—the sliding guides must support the window weight;
FR3—the power window system must fit the available space within the car door.
In particular, the present chapter will address FR1 as the user perceives the win-
dow stroke time as the main response of the system.
8.3.2 Building Object-Oriented Complete Model

of the Power Windows System
Simulation of a power window system could be accomplished using Object-

Oriented Modelling (OOM). Here, the modelling is accomplished by firstly pro-
viding a series of levels of detail as in Fig. 8.3. The power window system is
1A Bowden cable transmits mechanical force or energy through the movement of an inner cable
relative to an outer housing.

Fig. 8.3 Object-oriented model of power window system within Dymola environment
divided into three subsystems (first level decomposition) related to the control,
sensor and physical subsystems, respectively. Such decomposition is also gen-
erally used when tackling the design of mechatronic systems. Each subsystem
is then further decomposed according to functional groups using a second level
decomposition. Finally, the third level decomposition is related to generalised
components, components that characterize different physical properties such as
component parts or torque and force vectors or signals.
A primary need is evaluating of how physical components affect system behav-
iour. Therefore, the model related to the physical subsystem is taken into account.
As the power window behaviour is affected by the control system, the control and
sensor subsystems are subsequently modelled and simulated by using the set of
parameters defined for the physical subsystem.
The physical system is modelled using the Modelica language within the
Dymola environment. Six objects compose the model as follows:
• DC motor object—this is characterized by three parameters: resistance, induct-
ance and motor’s torque constant.2
• Gear object—this is characterized by two parameters: speed ratio and effi-
ciency. For a DC motor in a power window this is generally a worm gear with
an efficiency value of the order of 45 %.
• Inertia object—this represents the inertia of all rotational components.
• Pulley object—this is characterized by pulley radius.
2Fora DC motor it is possible to define kT i.e. the motor’s torque constant and kb i.e. the motor’s
back electro magnetic force (emf) constant. In SI units kT and kb are expressed, respectively, in
(N m)/A and V/(rad/s) and kT = kb.
Table 8.1 Design parameters for power window system

Design parameters
Gear Speed ratio Pulley diameter Inertia Efficiency
DC motor Resistor Inductor Motor torque constant Inertia
Sliding mechanism Inertia Friction
Window Weight Stroke
• Window object—this is modelled as a rectangular trapezoid without the dou-

ble curvature of the window. Only the magnitude in the window movement
direction is taken into account; window mass centre is computed by means of
Varignon’s Theorem [36].
• Mass object—this represents the window mass.
The power supply is modelled as a constant voltage generator of 13 V. At the
beginning, no control on the evolution of the system is taken into account. The
window stroke is equal to 0.40 m and is limited by means of source code and sim-
ulation stops when the stroke is completed. This allows the analysis of the system
evolution without control system effects. The model of the power window system
is shown in Fig. 8.3.
The model of power window system is characterized by a set of the design
parameters (Table 8.1). A change of design parameters entails the variation of
the lifting and/or lower performance as they produce the variation of a series of
related or dependent parameters.
8.3.3 Model Reduction by RSM
In order to build an empirically reduced model for the main system responses,
Central Composite Design (CCD) was adopted aiming at characterizing and opti-
mizing the system. A CCD is an experimental design often used in Response
Surface Methodology (RSM) to build a second-order (quadratic) model for the
response variable. It contains an embedded factorial design with centre points
which are augmented with a group of star points that allows an estimation of the
curvature [37]. Within the set of independent parameters we define following sub-
set of seven factors that are identified as the independent variable xi (for i = 1, 7):
A—Window weight (x1);
B—Speed ratio (x2);
C—Pulley radius (x3);
D—Inertia (x4);
E—Motor’s torque constant (x5);
F—Resistance (x6);
G—Inductance (x7).
Table 8.2 Levels of the factors (α = 1.63)

−α −1 0 +1 +α
A Window weight (kg) 2.00 2.77 4.00 5.23 6.00
B Speed ratio 55 65.59 82.50 99.41 110
C Pulley radius (m) 0.020 0.026 0.035 0.044 0.050
D Inertia (kg m2) 0.001 0.002 0.004 0.006 0.007
E Motor torque constant (Nm/A) 0.010 0.031 0.065 0.099 0.12
F Resistor (Ω) 0.400 0.997 1.95 2.90 3.50
G Inductor (H) 0.001 0.020 0.050 0.081 0.10
In particular, the Efficiency of gears and the Friction of the sliding mechanism
(see Table 8.1) are here considered as constant parameters. As required by CCD,
we determine factorial points (+1 and −1), centre points (0) and axial points (+α
and −α). According to the characteristics of power window systems available on
the market, it is possible to define the factor ranges in terms of alphas (+α, −α).
In particular, such factor ranges represent a set of designing choices that does not
imply significant cost variations. The values are shown in Table 8.2.
The full factorial plan is composed of 152 experimental runs. We carry out a
fractional factorial plan of 50 experimental runs, using 6 centre points and 44 non-
centre points. Here, the stroke time of the window Y[s] represents the FR1 of the
power window system to be assured. Therefore, stroke time Y is collected as main
response during the present analysis. The results related to 50 experimental runs
are briefly summed up in the following. In particular, the trend of stroke time is
represented when the factors change.
ANOVA analysis, performed for stroke time Y (Table 8.3), highlights that three
factors and one interaction are significant:
1. E—Motor’s torque constant;
2. C—Pulley radius;
3. B—Speed ratio;
4. CE—Pulley radius and motor’s torque constant interaction.
The model is significant and has an R-squared value of 0.91. The reduced model
is:
y = −3.295 + 0.061x2 − 10.08x3 + 137.9x5 − 2059x3 x5 (8.1)
Equation 8.1 can be used to predict the physical system response according to a
given set of values. In particular, it presents an interaction between two significant
factors (C and E). In the following, the prediction of system response, obtained by
means of the Eq. 8.1, is compared with the one coming from the simulation of the
power window system and obtained through the object-oriented model.
Table 8.3 Analysis of variance

Source Squares df Square Value Prob > F
Model 288.66 35 8.25 29.10 <0.0001
significant
A—window weight 0.54 1 0.54 1.91 0.1887
B—speed ratio 24.96 1 24.96 88.06 <0.0001
C—pulley radius 45.50 1 45.50 160.57 <0.0001
D—inertia (transmission) 3.026E-003 1 3.026E-003 0.011 0.9192
E—motor torque constant 127.52 1 127.52 449.99 <0.00011
F—resistor 2.74 1 2.74 9.66 0.0077
G—inductor 0.045 1 0.045 0.16 0.6957
AB 0.20 1 0.20 0.70 0.4158
CD 1.871E-003 1 1.871E-003 6.603E-003 0.9364
CE 8.91 1 8.91 31.43 <0.0001
CF 1.122E-003 1 1.122E-003 3.958E-003 0.9507
G^2 0.10 1 0.10 0.35 0.5619
Residual 3.97 14 0.28
Lack of fit 3.97 9 0.44
Total 292.63 49
Table 8.4 Data sets related to two system settings allowing a stroke time near to the target value
System setting n.1 System setting n.2
B Speed ratio 70 89
C Pulley radius (m) 0.025 0.019
E Motor torque constant (Nm/A) 0.041 0.021
A Window weight (kg) 3.4
D Inertia (kg * m2) 0.003
F Resistor (Ω) 1.5
G Inductor (H) 0.04
The non-significant factors are the same for the two settings
8.3.4 Improving the Robustness of the System
Equation 8.1 is here used to address a preliminary result in terms of stroke time
Y. Then, the simulation, performed through the complete object-oriented model,
is used to accomplish the tuning of parameters. Finally, by addressing the noise
factors it is possible to find the set of design parameters that makes the power win-
dow system robust.
By using Eq. 8.1 it is possible to identify two different settings of design
parameters that allow a stroke time close to the target value Y = 4 s. Table 8.4 con-
tains the data sets related to two system settings.
Table 8.5 Level assigned to noise factors

Setting n.1 Setting n.2
− + − +
NA Window weight (kg) 3.366 3.434 3.366 3.434
NB Speed ratio 69.3 70.7 88.11 89.89
NC Pulley radius (m) 0.02475 0.02525 0.01881 0.01919
ND Inertia (kg * m2) 0.00297 0.00303 0.00297 0.00303
NE Motor’s torque constant (Nm/A) 0.04059 0.04141 0.02079 0.02121
NF Resistor (Ω) 1.485 1.515 1.485 1.515
NG Inductor (H) 0.0396 0.0404 0.0396 0.0404
Table 8.6 Outer array for testing the setting n.1

Expt NA NB NC ND NE NF NG Y
# Window Speed Pulley Inertia Motor’s Resistor Inductor Stroke
weight ratio radius (kg * m2) torque (Ω) (H) time (s)
(kg) (m) constant
(Nm/A)
1 3.366 69.3 0.02475 0.00297 0.04059 1.485 0.0396 3.96
2 3.366 69.3 0.02475 0.00303 0.04141 1.515 0.0404 4.03
3 3.366 70.7 0.02525 0.00297 0.04059 1.515 0.0404 3.97
4 3.366 70.7 0.02525 0.00303 0.04141 1.485 0.0396 4.02
5 3.434 69.3 0.02525 0.00297 0.04141 1.485 0.0404 3.96
6 3.434 69.3 0.02525 0.00303 0.04059 1.515 0.0396 3.91
7 3.434 70.7 0.02475 0.00297 0.04141 1.515 0.0396 4.11
8 3.434 70.7 0.02475 0.00303 0.04059 1.485 0.0404 4.04
With reference to the system setting n.1, stroke time Y is equal to 4.27 s while,
by considering the system setting n.2 the stroke time Y is equal to 4.02 s. Then, by
using the object oriented model of the power window system the simulated stroke
time Y is equal to 4.0 s for both configurations. Therefore, the percentage error
between the reduced model (expressed by Eq. 8.1) and the complete object-ori-
ented model is 6.75 and 0.5 %, respectively.
It is then possible to apply the noise factors, as outer array, by taking into
account two different levels assigned to each noise factor (Table 8.5). The two dif-
ferent levels are evaluated by considering the mean value and then by subtracting
and adding the standard deviation. To simplify the analysis it is here assumed that
the standard deviation, evaluated for each parameter, is equal to the 1 % of the
nominal value.
Table 8.6 lists the outer array for the setting n.1, while Fig. 8.4 depicts the con-
tribution ratios around the setting n.1. It highlights the contribution of the variation
of each factor to the variation of Y. Similarly, Table 8.7 lists the outer array for the
setting n.2, while Fig. 8.5 depicts the contribution ratios around the setting n.2.
Fig. 8.4 Contribution ratios around the setting n.1
Table 8.7 Outer array for testing the setting n.2

Expt NA NB NC ND NE NF NG Y
# Window Speed Pulley Inertia Motor’s Resistor Inductor Stroke
weight ratio radius (kg * m2) torque (Ω) (H) time (s)
(kg) (m) constant
(Nm/A)
1 3.366 88.11 0.01881 0.00297 0.02079 1.485 0.0396 3.97
2 3.366 88.11 0.01881 0.00303 0.02121 1.515 0.0404 4.02
3 3.366 89.89 0.01919 0.00297 0.02079 1.515 0.0404 3.99
4 3.366 89.89 0.01919 0.00303 0.02121 1.485 0.0396 4.00
5 3.434 88.11 0.01919 0.00297 0.02121 1.485 0.0404 3.95
6 3.434 88.11 0.01919 0.00303 0.02079 1.515 0.0396 3.94
7 3.434 89.89 0.01881 0.00297 0.02121 1.515 0.0396 4.09
8 3.434 89.89 0.01881 0.00303 0.02079 1.485 0.0404 4.04
Fig. 8.5 Contribution ratios around the setting n.2

Table 8.8 Data related to performance indicators of the system

System setting # Mean value Standard deviation Average loss Signal-to-noise ratio
ym σ2 Lm SN
1 4.00 0.062 3.89 36.15
2 4.00 0.050 2.46 38.14
Starting from the collected data concerning the effects of noise factors, it is
possible to choose a more robust setting. The “nominal is the best” loss function
can be assumed as follows:
L(y) = k(y − y0 )2 (8.2)
where y0 = 4 and k = 1000. The average loss and Signal-to-Noise (SN) ratio could
be written as in the following:
Lm = k[(ym − y0 )2 + s2 ] (8.3)
2 2
SN = 10log(ym /s ) (8.4)
where ym and s2 are the mean and the standard deviation of the collected data,
respectively.
The performance indicators of the system could be evaluated using the data in
Tables 8.6 and 8.7. The results are reported in Table 8.8.
From these results, it is possible to choose the setting n.2 as the most robust set-
ting that can be assumed without cost increment. So the quality can increase with-
out increasing the cost, and this is the main aim of the RD methodology.
8.4 Conclusions
In the present chapter the issue of improving the robustness of mechatronic sys-
tems has been tackled. In particular, the need to operate at two levels (conceptual
and operational, respectively) has been highlighted in order to accomplish first
the mechatronic system architecture and, then, the mechatronic system design.
According to the improvement of robustness at conceptual level the criticality is
represented by the times to be spent, within iterative loops, to accomplish the diag-
onal form of design matrix, i.e. the set of decoupled design parameters.
According to the improvement of robustness at operational level, OOM and
related simulations demonstrated to be a valid tool for the evaluation of system
performances in presence of variability. Design of the experiments performed
for the main performance of a power window system is able to address the most
significant control factors. In this way, it is possible to locate the variations of
control factors allowing the conscious choose of design parameters, among
different system configurations.
This approach is addressed to the set of factors belonging to all subsystems

(physical-, sensor- and control subsystems) of a mechatronic system and it
is independent by its complexity. Therefore, even if further depths could be
accomplished, this approach appears a durable way to improve the designing of
such systems.
Acknowledgements Authors thank eng. Ferdinando Vitolo for his technical support in the
elaboration of data related to the automotive power windows system.
References
1. Bradley D, Seward D, Dawson D, Burge S (2000) Mechatronics and the design of intelligent
machines and systems. CRC Press, Taylor and Francis Group, Boca Raton, Abingdon
2. Bathelt J, Jonsson A, Bacs C, Dierssen S, Meier M (2005) Applying the new VDI Design
Guideline 2206 on mechatronic systems controlled by a PLC. In: International conference on
engineering design (ICED 05), Melbourne, Australia, pp 1–14
3. Zheng C, Bricogne M, Le Duigou J, Eynard B (2014) Survey on mechatronic engineering: a
focus on design methods and product models. Adv Eng Inform 28:241–257
4. Lefévre J, Charles S, Bosch-Mauchand M, Eynard B, Padiolleau E (2014) Multidisciplinary
modelling and simulation for mechatronic. J Des Res 12(1/2):127–144
5. Zupancic B, Sodja A (2011) Advanced multi-domain modelling: advantages and dangers,
UKSim. In: 13th International Conference on Modelling and Simulation, doi:10.1109/
UKSIM.2011.56
6. McBride RT (2005) System analysis through bond graph modelling, PhD thesis. Department
Electrical and Computer Engineering, University of Arizona
7. Borutzky W (2010) Bond graph methodology: development and analysis of multidisciplinary
dynamic system models. Springer, London
8. El Fahime B, Rzine B, Moujibi N, Saka A, Bouayad A, Radouani M (2010) Modeling,
simulation and analysis of mechatronic systems by bond graphs and monte carlo simulations.
J Model Simul Syst 1(4):2204–2212
9. Precup RE, Preitl S (2006) Stability and sensitivity analysis of fuzzy control systems,
mechatronics applications. Acta Polytech Hung 3(1):61–76
10. Bishop RH (2005) Introduction to mechatronics. In: Dorf RC (ed) The engineering handbook,
2 edn. CRC Press, Boca Raton
11. Bolton W (2009) Mechatronics: a multidisciplinary approach, 4th edn. Prentice Hall,
England
12. Lanzotti A, Vanacore A (2007) An efficient and easy discretizing method for the treatment of
noise factors in Robust Design. Asian J Qual 8(3):188–197
13. Lanzotti A (2008) Robust design of car packaging in virtual environment. Int J Interact Des
Manuf (IJIDeM) 2(1):39–46
14. Lanzotti A, Barone S (2008) Robust ergonomic virtual design. In: Erto P (ed) Statistics for
innovation: statistical design of “continuous” product innovation, Springer, Milano, pp 61–83
15. Gumus B (2005) Axiomatic product development lifecycle (APDL) model. PhD dissertation,
Texas Tech University
16. Yang K, El-Haik B (2003) Design for six sigma. McGraw Hill, New York City
17. Altshuler G (2000) The innovation algorithm: TRIZ, systematic innovation and technical
creativity (Shulyak L, Rodman S, (trans)). Technical Innovation Center, Worcester
18. Terninko J, Zusman A, Zlotin B (1998) Systematic innovation: an introduction to TRIZ
(theory of inventive problem solving). St Lucie Press, Boca Raton
19. Kaplan S (1996) An introduction to TRIZ: the Russian theory of inventive problem solving.
Ideation International Inc, Southfield
20. Savransky SD (2000) Engineering of creativity: introduction to TRIZ methodology of inven-
tive problem solving. CRC Press, Boca Raton
21. Myers RH, Montgomery DC, Anderson-Cook CM (2009) Response surface methodology:
process and product optimization using designed experiments, 3rd edn. Wiley, Hoboken
22. Taguchi G (1987) System of experimental design. Kraus International Pub, New York
23. Phadke SM (1989) Quality engineering using robust design. Prentice-Hall International Ed,
Upper Saddle River
24. Park S (1996) Robust design and analysis for quality engineering. Chapman & Hall, London
25. Wu CFJ, Hamada M (2000) Experiments. Wiley, New York
26. D’Errico J, Zaino RA (1988) Statistical tolerancing using a modification of Taguchi’s
method. Technometrics 30(4):397–405
27. Pignatiello J (1993) Strategies for robust multiresponse quality engineering. IIE Trans
25(3):5–15
28. Spiring FA, Yeung AS (1998) A general class of loss functions with industrial applications.
J Qual Technol 30(2):152
29. Kuhnt S, Erdbrügge M (2004) A strategy of robust parameter design for multiple responses.
Stat Model 4:249–264
30. Murphy TE, Tsui K-L, Allen JK (2005) A review of robust design methods for multiple
responses. Res Eng Des 16:118–132
31. Johansson P, Chakhunashvili A, Barone S, Bergman B (2006) Variation mode and effect
analysis: a practical tool for quality improvement. Qual Reliab Eng Int 22(8):865–876
32. Bae SJ, Tsui K-L (2006) Analysis of dynamic robust design experiment with explicit and
hidden noise variables. Qual Technol Quant Manage 3(1):55–75
33. Maghsoodloo S, Li MC (2000) Optimal asymmetric tolerance design. IIE Trans
32:1127–1137
34. Lanzotti A, Patalano S (2002) A CAT-based approach to optimum tolerance allocation. In:
Proceedings of the 3rd CIRP ICME seminar, Ischia, Italy, pp 403–408
35. Seo HS, Kwak BM (2002) Efficient statistical tolerance analysis for general distributions
using three-point information. Int J Prod Res 40(4):931–944
36. Coxeter HSM, Greitzer SL (1967) Geometry revisited, mathematical association of America,
ISBN 0-88385-600
37. Box GEP, Hunter JS, Hunter WG (2005) Statistics for experimenters: design, innovation, and
discovery, 2nd edn. Wiley, New York City
Chapter 9
Integrated Manufacturing: The Future
of Fabricating Mechatronic Devices
Nicholas Fry, Rob Richardson and Jordan H. Boyle
9.1 Introduction
This chapter explores current and future technologies for physically creating
mechatronic devices, and in particular robotic systems. Robots consist of electron-
ics, actuators and sensors within a self-contained mechanical structure and have
the ability to exert controlled external forces to enable them to physically inter-
act with the world around them. There is no doubt that robots have the potential
to revolutionise many sectors [1], but there are many barriers to widespread use
including public perceptions and difficulties in physically and computationally
integrating robots into real-world environments. The cost of both designing and
manufacturing robots is also very high. Improving manufacturing techniques for
robots and providing better integration between the mechanical and electrical sys-
tems could help robots become physically robust, small, sealed, mobile and appro-
priate for the many challenging environments where their use could have a big
impact.
9.2 Traditional Manufacturing Methods
Conventionally, mechatronic devices are created from discrete electronic, actuator,

sensor and computation modules, integrated via a mechanical structure using bolts
and other fastenings. The use of fastenings to secure all the parts limits robust-
ness, as anything that has been assembled can also be disassembled [2]. The joints
between components are common weaknesses, and as the device gets smaller the
N. Fry · R. Richardson (*) · J.H. Boyle

University of Leeds, Leeds, UK

DOI 10.1007/978-3-319-32156-1_9
130 N. Fry et al.
space taken up by these fastenings becomes increasingly significant. Assembly of

these complex devices is also a big challenge; once a product has been designed,
it must undergo a rigorous ‘design for assembly’ process to make it suitable for
manufacturing. This often introduces large costs and delays and greatly limits the
designer. For example, the shape of a mechanical case must be simple enough that
it can be released from a mould; circuit boards (both rigid and flexible) can only
be populated when held flat and must incorporate off-the-shelf sensors in specific
packages (physical cases) and footprint sizes (the size and shape of the compo-
nent connector). Manufacturing processes dictate the way we design and produce
items, imposing limitations on what is possible. We predict that advances in
Additive Manufacturing (AM) will enable a new Integrated Manufacturing method
that allows fabrication around mechatronic components. This could remove the
need for post-fabrication assembly and the use of fastenings, yielding rapid pro-
duction of robust devices.
9.3 Current Methods: The Rise of Direct Digital

Manufacturing
Traditionally manufacturing was either manual, making it slow and labour inten-
sive, or automated by simple robot operations, making it inflexible and costly to
set up—requiring large product numbers to justify the expenditure. Computer
control and digital design are helping to change the economic patterns that have
defined manufacturing since the industrial revolution.
9.3.1 Automation
The design of a production line revolves around the need to make it as efficient
as possible. Adjacent manufacturing operations should be located as physically
close as possible and the movement of human workers and parts carefully choreo-
graphed to minimise wasted actions and time. While the philosophy behind assem-
bly lines has not changed since they were introduced, the technology that powers
them has, and it brings along its own set of advantages and challenges.
Through the application of sensors and actuators factories are becoming
increasingly more automated. Where once a human worker had to load parts into
a machine, a robot arm can now be coupled with a computer vision system that
identifies a specific part and places it in the jig itself. This can increase the speed
and efficiency of the process and is often safer, removing the need for a human to
work closely with dangerous machinery.
For low volume manufacturing, Computer Numerically Controlled (CNC)
machines are arguably having an even greater effect. The first CNC machines were
9 Integrated Manufacturing … 131
traditional machine tools, such as milling machines and lathes, which were con-
trolled by a computer. They have now expanded to include laser cutters, 5-axis
milling machines and even 3D printers. CNC gives huge advantages as machines
can quickly take a digital version of a part and turn it into a very high precision
physical model. This disrupts the highly manual process that was traditionally
used for small production runs, reducing the time and cost to produce parts while
increasing the quality and precision. There is also an increased flexibility from the
automation of these machines as changing the digital design file, changes the part
being produced.
9.3.2 Additive Manufacturing
Additive Manufacturing (AM) is a disruptive technology that was first developed

just over 30 years ago, but has burst into the public consciousness in recent years,
as consumer desktop 3D printers become cheaper and more readily available.
Additive Manufacturing and 3D printing are interchangeable terms to describe a
range of technologies. The American Society for Testing and Materials (ASTM)
International defines AM as “A process of joining materials to make objects from
3D model data, usually layer upon layer, as opposed to subtractive manufacturing
methodologies” [3].
There are different methods of depositing material in AM processes, but in all
cases the first step is to digitally slice the computer model into thin layers. The
layers are then physically laid down, one after the other, gradually building up a
three-dimensional part. Thinner layers give finer detail, but also increase print time
as more layers are required. Deposition technologies can be split into four broad
categories [4], although the ASTM breaks these down further [3].
First there are extrusion-based machines. These feed semi-liquid material out
of a moving nozzle, which then sets to form a hard layer. Many different materi-
als have been used, including clay and chocolate, although thermoplastics are by
far the most common, especially in consumer-grade 3D printers. These are com-
monly referred to as Fused Deposition Modelling (FDM) printers, although this is
a trademark of Stratasys®. Various acronyms have been coined, but we will refer
to this process as Fused Filament Fabrication (FFF).
Second, there are techniques that bind particles of powder together, either using
glue or a heat source (e.g. a laser). In this method the part is constructed inside a
vat of powder. To form a layer, an arm first sweeps a thin film of powder across the
top of the print bed. The print head then traces out the contours of that layer and
solidifies the powder.
Third is photopolymerisation, a term which refers to solidification of a liquid
resin using a light source. Lasers, UV lights and even digital projectors can be
used. Stereolithography (SLA or SL) was the first form of AM to be developed,
and is widely used in industry due to its high resolution. An SL machine includes
a vat of resin with a perforated build plate inside. At the start of a build the plate is
132 N. Fry et al.
Fig. 9.1 Effect of layering, a subtractive machining yields a smooth surface, b AM part has a
stepped surface and requires support material under the overhang
positioned just below the surface of the resin while a laser traces out the layer, cur-
ing the resin. The plate then moves down by the height of one layer and the resin
either flows, or is wiped across, before repeating the process.
Finally there is sheet lamination, a process in which each layer is formed by
cutting the shape from a thin sheet of paper, plastic or even metal. These sheets
are bonded together using an adhesive or ultrasonic welding. The paper-based
approach produces parts with poor mechanical properties, but can be combined
with inkjet printing to give full colour models.
The layer-by-layer construction technique that is common to all these processes
enables a new freedom in design as very complex shapes; even parts that cannot
be manufactured using traditional techniques can often be realised using 3D print-
ing (e.g. nested spheres). However, there are some limitations to the geometry
of the parts that can be manufactured. The use of layers can cause a ‘stair step’
effect on curved surfaces (Fig. 9.1) as the shape must be digitised into discrete
layers. Furthermore, each layer must be supported by a layer underneath, which
means that any overhanging sections must be supported. Support structures are
either printed in the same material as the part and designed to be cut, ‘snapped’
or machined off, or a different material is used that can be washed, peeled or dis-
solved away.
Although the impact of AM technology is sometimes exaggerated, it has unde-
niably had a dramatic effect on the way products are manufactured and will con-
tinue to do so. Since the Industrial Revolution, Economies of Scale have defined
the structure of the manufacturing industry [5]. When a greater number of prod-
ucts are produced, cost per unit decreases. Unfortunately, this requires a large ini-
tial investment for manufacturing to become cost effective and therefore leads to
large-scale, centralised, mass manufacturing. Due to the fact that AM requires no
tooling or set-up, the price per unit levels off very quickly, and is often modelled
as a flat line when compared with traditional manufacturing cost structures [6]. As
Fig. 9.2 shows, this makes AM a very cost effective option for low-volume manu-
facturing. Currently traditional methods are more cost effective for large quanti-
ties, but the break even point will improve as 3D printers become more capable,
cheaper and faster.
Fig. 9.2 A comparison

of traditional and additive
manufacturing cost models
AM will also significantly impact the Economies of Scope. This term refers to
the fact that if one set of equipment, processes or materials can be used to make
different products then the unit cost falls as the investment is split between more
items [5]. 3D printers are extremely versatile in the geometries they can produce
and there is little time or cost penalty incurred when swapping between parts.
They can even produce different objects side by side on the build plate.
The abilities of 3D printers are so different from traditional manufacturing that
they may allow new approaches to production that are not practical or possible
with current manufacturing methods. For example, the layer-based process allows
very complex geometries that could not be machined or cast due to object geom-
etry limiting the access for tools or mould removal. Internal features can also be
created, without the need to split the part into many pieces. This in turn leads to
parts that can be lighter, require less assembly and have complex geometries, all
while reducing material waste and lead times.
Additive Manufacturing was originally used exclusively as a Rapid Prototyping
(RP) technology, as it allowed product designers to produce a physical model of
their design within hours rather than days or weeks. Now, however, AM is expand-
ing into Direct Digital Manufacturing (DDM). This is where AM technology is
used to manufacture high quality, final production parts, rather than just proto-
types. While the majority of 3D printers available are still aimed at RP, we are cur-
rently on the cusp of this exciting shift, where AM is being used to shorten supply
chains, produce optimised parts and manufacture custom components. In fact, it
is estimated that about 20 % of 3D printed parts are already for end use. Within
5 years it is predicted that this will be the majority, producing industrial tooling,
individual components or full final products [4].
The rest of this chapter will focus on currently emerging technologies that will
unlock the full potential of AM to revolutionise mechatronics manufacture. The
far-reaching consequences of this revolution are explored in the final section.
134 N. Fry et al.
9.4 Future Technology
9.4.1 Embedding Parts: Towards Non-assembly Structures
Joining components with fastenings tends to impose a set way of building

mechatronic devices which relies on specific materials and fastening techniques
and often results in a design which is inherently limited in terms of size, geometry
and robustness [2]. The layer-by-layer construction of AM gives the unique oppor-
tunity to access the enclosed internal geometry of a part. This gives rise to the
possibility of embedding components (motors, sensors, electronics) within object
structures—to become an inherent part of the structure. This approach can confer
major benefits when creating complex mechatronic systems.
Consider a humanoid robotic hand, for example. Researchers built two ver-
sions, the first of which used traditional assembly and included 60 parts (40 of
which were fastenings) in each finger [7]. For the second version each finger was
a single part manufactured using a hybrid AM method called Shape Deposition
Manufacturing (SDM). SDM is a layer-based manufacturing method which
involves a cycle of depositing material followed by precision shaping using sub-
tractive methods. This approach allows different materials to be incorporated in a
part, and even for components to be embedded inside the structure. Sensors were
integrated into each finger, creating a reliable interface while protecting the sensi-
tive electronics [7]. The joints and finger pads are made of a compliant material,
while the link sections are a stiff polymer. This compliance greatly increases the
robustness of the gripper; in fact the researchers have released a video of it being
repeatedly hit with a baseball bat, with no resulting damage [8]. Researchers also
selected the SDM process to mimic the impressive agility of a cockroach [9] and
the climbing of a gecko [10].
The ability to make inherent joints is a key to simplifying assembly [11] and
allows unique joints to be created. For example, the process has enabled ball and
socket joints to be redesigned to limit their motion in certain directions by chang-
ing their internal geometry—something that is not possible using manufacturing
methods other than AM [12].
Producing non-assembly joints (working mechanisms that require no assem-
bly) is difficult; support material typically has to fill small gaps between parts [11].
Small gaps are required to give good tolerances for the joints, but this makes the
support removal problematic. Traditional joints have been redesigned to result in
close tolerances and simple support removal [11], but this work must be extended
to enable a wider range on non-assembly joints.
Embedding inserts during the 3D printing process has given some interesting
results, with components such as motors, circuit boards and gears being inserted to
create, for example, a moving model of the stereolithography machine that printed
the parts [13], a remote controlled buggy [12] and a robotic hand [14]. These
inserts could also change the mechanical properties of the object. For example,
metal could be inserted to increase the stiffness of a part.
Fig. 9.3 Use of inserts in 3D printing, a print head collides with protruding insert b Insert
cannot be embedded until the build is as tall as the insert, but by that point the cavity could be
enclosed
There are challenges to overcome, however. Existing 3D printers are designed

to work with planar horizontal layers, so the top of the part being built must
always be flat to avoid collisions with the print head or recoating wiper arm (see
in Fig. 9.3a). This requirement prevents the insert from being embedded until the
build level is as high as it is. However, if the upper geometry of the part is convex,
then the cavity will be enclosed by the time the build level is high enough to shield
the component (Fig. 9.3b).
To overcome these issues, Shape Converters can be used (Fig. 9.4). These are
additional parts that fit around the component, converting its geometry into one
with vertical sides and a horizontal top that can be inserted without protrud-
ing [15]. These allow embedding but remove some of the possible advantages by
adding extra process planning and assembly steps. The shape converter must be
designed (it could have a complex shape) and manufactured, before being manu-
ally attached to the component for insertion into the main build. In more recent
work, the ability to print using transparent materials has enabled printing around
embedded optical elements to create displays and illumination, or even custom-
ised optoelectronic sensors [16]. This required infrared emitters and receivers to
be embedded during the printing process. These were embedded by pausing the
printer and manually inserting the component and a shape converter; the same
labour-intensive technique that was developed a decade before [13]. The creation
of fully integrated robotic mechanisms is being held back by a lack of suitable
embedding processes that allow inserts to be seamlessly built around.
136 N. Fry et al.
Fig. 9.4 A shape converter is assembled with the insert, converting the geometry to vertical
sides and horizontal top, removing collisions
9.4.2 3D Printed Sensors
Including sensors, actuators and electronics in 3D printed structures will turn them
into functional mechatronic devices. There is little reason to 3D print versions of
parts that are easy to acquire and could simply be inserted, yet the integration and
customisation potential of printing certain components does bring unique advan-
tages that will be valuable in many cases. For example, a method to print speakers
in any shape has been developed [17]. This allows the component to be totally
integrated with the device it is in, allowing much greater design freedom, as the
casing no longer needs to fit the shape of a standard speaker cone. By changing the
shape of the speaker, the focus of the sound can also be adjusted [17]. A cylindri-
cal speaker could be used to make an ultrasonic sensor capable of detecting obsta-
cles all the way around a robot. Or a single object could have many individual
speakers giving very directional sounds which could be used for artistic displays
or communication [17].
The ability to print soft and transparent structures allowed researchers to
develop novel optoelectronic sensors [16]. The electrical components can be situ-
ated near the surface of the part and externally wired to a control circuit, while
sensing can be achieved using light signals channelled through printed optical con-
duits, avoiding the need for conductive material within the part.
By embedding infrared (IR) emitter and receiver pairs into a part, many dif-
ferent sensors can be made [16]. Using an Objet multi-material 3D printer, flex-
ible, clear material can be used to guide the light [16]. Deforming this light guide
changes the reflections, allowing movement to be detected. Switches, sliders,
encoders and accelerometers have all been made using different configurations,
for example a button is made by guiding IR light from the emitter to the receiver
with a cantilever beam. When the user presses the button down it deforms the light
guide, bending it away for the receiver. The receiver now registers a much lower
signal and can be used to trigger a different electrical circuit. Using this technique
customised, accurate sensors and user interface devices can be easily, quickly and
repeatedly incorporated into any product [16].
9.4.3 Integrating Multiple Materials
Extending 3D printing processes to work with multiple materials opens up a huge

range of possibilities for the technology, and while it is available for some meth-
ods, there are still challenges to overcome before this becomes a mainstream man-
ufacturing technique.
The ability to combine different materials into one part would allow the
mechanical and aesthetic properties to be adjusted. This can increase functional-
ity as compliance could be included in different sections, or different colours or
surface finishes could make AM more appropriate for final production runs as
opposed to prototypes [18]. If materials with special properties, such as electrical
conductivity, could be incorporated, then we would be much closer to an all-in-
one process to fabricate functional mechatronic devices.
The Stratasys® Objet range of 3D printers are probably the best multi-material
printers available at the time of writing [19]. They can be used to print parts with
combinations of flexible, hard, transparent and full colour materials. The printer
uses the Polyjet process, somewhat similar to that of an inkjet printer, in which
tiny droplets of liquid build material are deposited from holes in the print head.
The print head also includes an ultraviolet light that passes over each layer to cure
it and solidifies the polymer before laying down the next layer. This jetting process
means that different materials can be combined in a controlled way to create func-
tional gradients of materials. In a process such as SDM each material is deposited
individually with a clear interface between the sections. In contrast, Polyjet allows
gradual material transitions that remove the weakness caused by a sudden change
in materials. The material properties can even be precisely adjusted at differ-
ent points throughout a model. This is an ability that is only possible with multi-
material 3D printing. Adjusting the compliance, elasticity and damping of a part
through its structure could have applications in many fields. For example, surgical
devices may need to be strong and stiff in one section to aid handling or insertion
while having a soft-end effector to reduce tissue damage. Achieving this with one
integrated part would also facilitate waterproofing and sterilisation. In robotics,
flexible joints actuated by shape memory alloy wire have been created, showing
the potential of integrated non-assembly joints [20].
The Polyjet process, while giving many advantages, also limits the materials
that can be used. As the polymers are cured using UV light, each material must be
created to work in the printer, giving rise to Stratasys’® ‘rubber-like’ and ‘ABS-
like’ descriptions of its materials. They share similar properties but are not the
138 N. Fry et al.
same as the traditional materials engineers are familiar with using. One of the big-
gest challenges, and opportunities, for 3D printing is increasing the range of mate-
rials that can be reliably used.
Malone and Lipson created the Fab@Home project, in which they designed a
multi-material 3D printer that is open source and low cost [21]. They used this
to demonstrate that using different materials and processes during a build they
can produce working electromechanical devices. The system uses a syringe to
deposit a wide range of liquid, gel or paste materials and a molten extrusion head
to deposit thermoplastics and solder. This enabled the creation of flexible circuit
boards, strain sensors, electromagnets, electromechanical relays, electroactive
polymer relays and even batteries in different shapes. While creating functional
electromechanical parts is a huge step for additive manufacturing, the Fab@Home
system is low cost, low resolution and labour intensive, as processes must be care-
fully planned and the various materials are manually supplied to the machine.
These components were functional, but cannot match the performance of their tra-
ditionally manufactured counterparts. Process planning is a difficult step as multi-
material additive manufacturing gives new ways of making parts, so no set way
of working has been established. This will continue to be challenging as different
technical processes will affect the design and planning stage.
9.4.4 3D Printed Electronics
One of the most keenly awaited developments in AM is the ability to print electron-
ics and conductive traces. This would enable functional mechatronic devices to be
directly manufactured in one process. Novel designs that are not constrained by the
need to include a flat-printed circuit board would be possible. Wires could also be
replaced with conductive tracks, removing the difficulty of routing wires through
complex and tight spaces and reducing the size and weight of the resulting device.
Batteries and transistors have been made via 3D printing [21], but there is a
long way to go before these parts can compete with the efficiency and price of
established processes. In the future many electrical components may be printed
from their basic materials, allowing a huge catalogue of parts to be used without
the need for each to be kept in stock. Despite this, merely printing conductive con-
nections and inserting premade electronic components would still open up a huge
range of possibilities. For the foreseeable future, at least, this will be the preferred
method of incorporating electronics into an AM part as the mass manufacturing
processes for electronic components are so well optimised.
The main options for printing conductive connections are: depositing con-
ductive inks; extruding solder; extruding conductive thermoplastic filament or
embedding wires. Commercial aerosol jet printers are available and have been
demonstrated to produce a circuit on the wing of a UAV Model constructed via
3D printing [22]. The process uses a mist generator to atomise conductive ink,
then aerodynamically focuses it, using a sheath gas, to create a fine flow. Feature
sizes less than 15 µm are possible and the approach shows great promise for high-
resolution circuits [22].
Conductive ink is used in a different way in the Voxel8 3D printer [23]. This is
a high-end consumer 3D printer that combines FFF and pneumatic ink dispensing
to create the first true embedded, 3D, electronics printer. The mechanical system is
not particularly impressive here. Rather it is the material and control advances that
are the keys. First, the ink they have created is an order of magnitude more con-
ductive than others and, importantly, cures at room temperature. Second, through
a partnership with Autodesk® they are offering a Computer Aided Design (CAD)
program that enables circuit traces to be designed in 3D and for the printer to
pause at the correct time to allow components to be inserted. Methods for process
planning are still required, while design rules and best practice for laying out cir-
cuits in three-dimensional space will need to be explored.
Embedding wires is an alternative method of incorporating conductive paths
into the AM build process and is appealing due the potential of low resistance,
multicore or shielded wires, along with low cost. Wire is difficult to embed,
however, as it does not directly adhere to the object and needs to be in tension
to manipulate. Wire has been successfully embedded into a thermoplastic part by
heating the surrounding material to soften it and pushing the wire just below the
surface. Few technical details are available, however, and the effect on the surface
finish has not been shown [24].
At a consumer level, FFF printers are the most common so there is great inter-
est in conductive materials for this process. Researchers have used low melting
point alloys, in the form of solder, as filament for FFF printers, which allows cir-
cuits to be printed [25]. This is cheap, readily available and can be used in unmod-
ified FFF printers. However, it is difficult to control the feeding and cooling rates
to give good results. There is much excitement around new conductive filaments
that can be used in FFF printers. The filaments are made by mixing a thermoplas-
tic, such as PLA with a conductive material. Graphene and nanocomposite materi-
als have been shown to provide the best results but the resistance is still relatively
very high. If short traces and low currents are used then functional devices and
circuits can be successfully created using this method. An Arduino light sensor
shield and single part flashlights have been printed, a simple robot has also been
designed and will shortly be printed [26].
The fact that this conductive material can be deposited with the same process as
the main build material simplifies the task of embedding the tracks within the part
and means many printers will require no custom hardware. This will greatly speed
the process of adoption. While these conductive materials are far from ideal and
the resolution of FFF does not approach that of Aerosol jet or standard PCBs, the
availability of this process to a large and inventive community will drive a rapid
push in innovation as individuals attempt to solve a wide variety of problems and
use it to manufacture their own designs. This will be unlikely to pose a threat to
companies interested in using 3D circuits for a competitive edge; however they
would be wise to pay attention to this unofficial research, as it is likely to be where
novel uses and applications are explored.
140 N. Fry et al.
9.5 Future Impact
While many individual experiments have confirmed the great potential for using
AM to manufacture end-use mechatronic devices, combining these technologies is
yet to happen. This section will look at the possibilities opened up by technologies
such as embedding components and printing electronics along with challenges that
engineers will need to face to realise them.
9.5.1 Fully Integrated Mechatronic Devices
If discrete components could be combined with multiple materials and conduc-

tive tracks, entirely new types of mechatronic devices could be created. Traditional
motors and bearings could be inserted alongside integrated AM compliant joints,
adding damping and robustness. Conductive tracks could connect actuators to a
3D printed battery whose shape has been customised to fit in the gaps between
other components, allowing the device to be smaller. The conductive tracks could
follow complex paths around the structure of the material, removing the need for
a flat PCB and allowing wires to be protected. These tracks may then lead to a 3D
printed socket allowing simple connections to different modules.
All of this would be integral to the mechanical structure of the device.
Assembly would happen during manufacture, eliminating joints in the casing,
clips that can snap or screws that could come loose. Removing the need for assem-
bly could speed up manufacturing, reduce labour costs and help provide consistent
quality. The monolithic structures that could result from fully integrated manu-
facturing would be ideal for harsh environments where toughness or sealing is
important.
One must consider the whole life cycle of a product, however. It may be neces-
sary to include access panels for maintenance and repair. Traditional inspection
methods may be impossible so new processes will be required. Tight integration of
multiple components and materials will present a particular challenge at the end of
the product’s life when the materials should be recovered and recycled. Some AM
processes use materials such as thermoplastics that can be remelted, but others use
chemical reactions that are not as simple to reverse.
While inserting components that are made in other ways is likely to be the most
efficient method in the near future, it is possible that some components will ben-
efit from 3D printing. Custom sensors, for example, may be a good candidate for
Additive Manufacturing since modifying the shape to fit exactly where they are
needed could be very useful. Or it may be that integrating basic but cheap sensors
is best done through AM, while more complex parts are manufactured in an exter-
nal process and embedded. This trade-off between the benefits of different manu-
facturing processes will have to be closely examined for each component to find
the most efficient process.
9.5.2 Additive Mass Manufacturing
Mass manufacturing takes advantage of the economies of scale and allows large
numbers of items to be made quickly and cost effectively. This is vital for compa-
nies who want to sell to large markets but this one-size-fits-all approach can result
in a product which is not ideal for anyone. Additive manufacturing on the other
hand takes longer to build an item but allows each unit to be personalised with
almost no limits to the customisation available. It also allows geometries that can-
not be made in any other way, opening up new possibilities for consumer products.
The on-demand nature of AM is also a huge advantage. As businesses and con-
sumers push for shorter and shorter lead times, the ability to speed up the produc-
tion cycle has benefits in many sectors.
Combining the advantages of mass and additive manufacturing would create
a totally new consumer environment. The way products are designed, fabricated
and consumed would change. Products with mass appeal could be sold across the
world; each one having the same underlying features, but with modifications to
reflect the needs of the local market, or even the individual consumer.
Customisation not only affects the function and appropriateness of a product,
but it can also add value emotionally, helping to increase the product’s appeal to
the user. Additionally, smaller markets that were once economically unfeasible
could now be developed for, broadening the range of products available and giving
manufactures wider market shares [27].
Google recently partnered with 3D Systems to develop a high-speed additive
manufacturing machine to use for Google’s modular smart phone Project Ara
[28]. The machine 3D Systems have developed looks more like a mini production
line than a traditional printer. It features 16 static print heads above a ‘race track’
where multiple build plates pass the heads at high speeds. This allows the machine
to be able to deposit up to 4 billion droplets of build material in 1 minute, which
made it 50 times faster than any other 3D printer at the time [28]. Unfortunately,
the technology was not mature enough for Google to deploy so they are currently
using a different method. Such abilities are still very much in demand however.
The Netherlands Organisation for Applies Scientific Research (TNO) has been
developing a similar system that features 100 small platforms moving at 2 m per
second under high-precision inkjet heads [29]. TNO have gone a step further than
3D systems and aim to incorporate other production processes including pick-and-
place robots and surface-finishing equipment. This allows components to be inserted
during the build process and for the system to make fully finished products. TNO
claim that this facilitates the move from prototyping to manufacturing for AM.
While mass additive manufacturing will surely be possible in the near future,
the systems discussed are still pilot systems. To convince manufactures to invest
in this technology they must clearly demonstrate speed and cost benefits over cur-
rent methods. Also, designing the machines to integrate seamlessly with current
production lines and equipment is vital. Companies will not change their whole
production method and equipment inventory just to facilitate the use of new
142 N. Fry et al.
technologies. When using AM for prototypes quality assurance and reliability are
not so important, but for finished products the manufacturer must be confident
that the process will be consistent and reliable, always producing parts that fit the
specification. Despite the early stage of these systems it is inevitable that Direct
Digital Manufacturing will eventually be used to produce a range of products, let-
ting manufacturers respond quickly to consumer demand.
9.5.3 Designing Differently: The Integrated Design Process
Fully integrated manufacturing methods will require new design processes.

Additive Manufacturing enables part geometry that could not be made any other
way, be it moulded or machined, and frees the designer from considering the com-
plexities of multi-part moulds or tooling requirements. However, the procedure is
not yet as simple as just pressing print. The process planning step is still vital for
checking if the design is suitable for AM. Currently designers often try to mini-
mise the support structures needed to build a part, but the introduction of multiple
materials and components will add further considerations. Engineers and design-
ers must ensure there is a good dialogue as these new AM technologies are devel-
oped, ensuring that the effects of any design for manufacturing rules are reduced.
Separate CAD packages are currently required for mechanical and electrical sys-
tems but this will need to change. It is important that user focused, intuitive soft-
ware is developed alongside new integration technologies to enable the designer to
consider a holistic mechatronic design.
Communication between the CAD package and the printer is also important.
Since the mid-80s when the STL was developed by 3D Systems it has been the
standard file format for rapid prototyping and AM systems. This specifies the
surface geometry of a 3D object, but does not include any information on the
materials, colours or textures used. This information is no longer sufficient for
modern multi-material printers. To address this issue the ASTM has developed the
Additive Manufacturing File Format (AMF). This is backwards compatible with
STL, but also adds features such as being able to specify multiple materials, col-
our and geometry with higher fidelity. 3D printer manufacturers and CAD soft-
ware developers must now take the lead and add compatibility to their products.
The ability to create a streamlined workflow from conception to fabrication will be
important for integrated design.
9.6 Conclusions
Additive manufacturing is making a huge impact in the manufacturing industry.

However, only when it is combined with embedded parts, multiple materials and
integrated electronics with its full potential to improve mechatronic and robotic
devices be realised. AM removes the need to rely on the economy of scale and
immediately gives a good economy of scope when manufacturing devices. Soon
customisation will be available in a huge range of products and industries, which
will be particularly important for robotics. There are many jobs that would benefit
from automation and the use of robots, but creating a single mass market robot
platform to meet these needs is still unrealistic, and small numbers of devices are
expensive to build, limiting the market for robotics. If an enhanced 3D printing
machine could be created that included abilities such as embedding functional
components, printing joints and connecting electronics then customised robots,
designed for specific applications would be much more readily available. This
method of construction could also make the robots more robust which is extremely
important for moving robots from laboratory-based experiments to real-world
applications.
The process is also made cheaper and simpler, by minimising the assembly
required, reducing material waste compared to subtractive methods and reduc-
ing the skills operators require. As 3D printing does not require long or expensive
set-up, the lead times for parts could be greatly reduced. Combined with minimal
assembly, on-demand manufacturing could be utilised in new applications such as
in surgery (custom implants or tools could be created), or in search and rescue
(a robot could be produced to fit a particular void or move on a specific surface).
Combining these innovations could also be the next step required for self-repli-
cating 3D printers, such as the RepRap project [30]. Currently, these printers can
produce 50 % of their own parts (not counting nuts and bolts). Using the ideas of
Integrated Manufacturing these fastenings could be removed and the progress in
printing electronics may soon help these machines reach the goal of 100 % print-
able parts.
While it is clear that there is great potential for Additive Manufacturing, many
challenges have been identified. Improved methods for embedding parts are a criti-
cal area that needs development, and printing electrical connections reliably with
low resistance will be vital for mechatronic and robotic devices. Inherent issues
such as discreet layers and the requirement for support material also need to be
mitigated. Finally, combining all the technology together, with intuitive software
and established process planning methods will finally bring about Direct Digital
Manufacturing and Integrated Mechatronic Manufacturing.
References
1. Willetts D (2013) Eight great technologies. Government Report

2. Bailey SA, Cham GJ, Cutkosky MR, Full RJ (2000) Biomimetic robotic mechanisms via
shape deposition manufacturing. In: International symposium robotics research. www.cdr.sta
nford.edu/biomimetics/pdf/isrr1999_bailey.pdf. Accessed 28 Dec 2015
3. ASTM F2792-12a (2012) Standard terminology for additive manufacturing technologies.
ASTM International, West Conshohocken, PA
4. Barnatt C (2014) 3D printing, 2nd ed. CreateSpace Independent Publishing Platform
144 N. Fry et al.
5. Cotteleer MJ (2014) 3D opportunity: additive manufacturing paths to performance, innova-

tion, and growth. Deloitte Review
6. Ruffo M, Tuck C, Hague R (2006) Cost estimation for rapid manufacturing—laser sintering
production for low to medium volumes. Proc IMechE Part B J Eng Manuf 220:9
7. Dollar AM, Howe RD (2006) A robust compliant grasper via shape deposition manufactur-
ing. IEEE/ASME Trans Mechatronics 11:2
8. Guizzo E (2012) iRobot smashes its new robotic hand with baseball bat. http://
spectrum.ieee.org/automaton/robotics/robotics-hardware/irobot-smashes-its-new-robotic-
hand-with-baseball-bat. Accessed 28 Dec 2015
9. Cham JG, Bailey SA, Clark JE, Full RJ, Cutkosky MR (2002) Fast and robust: hexapedal
robots via shape deposition manufacturing. Intl J Robot Res 21(10–11):869–882
10. Cutkosky MR, Kim S (2009) Design and fabrication of multi-material structures for bioin-
spired robots. Philos Trans Roy Soc A Math Phys Eng Sci 367:1894
11. Chen Y, Zhezheng C (2011) Joint analysis in rapid fabrication of non-assembly mechanisms.
Rapid Prototyping J 17:6
12. De Laurentis KJ, Kong FF, Mavroidis C (2002) Procedure for rapid fabrication of non-
assembly mechanisms with embedded components. In: ASME 2002 international design
engineering technical conference and computers and information in engineering conference,
Montreal
13. Kataria A, Rosen DW (2001) Building around inserts: methods for fabricating complex
devices in stereolithography. Rapid Prototyping J 7(5):53–62
14. De Laurentis KJ, Mavroidis C (2004) Rapid fabrication of a non-assembly robotic hand with
embedded components. Assembly Autom 24(4):394–405
15. Liao YS, Li HC, Chen MT (2007) The study of rapid prototyping process with embedded
functional inserts. J Mater Process Technol 192–193:68–74
16. Willis K, Brockmeyer E, Hudson S, Poupyrev I (2012) Printed optics: 3d printing of embed-
ded optical elements for interactive devices. In: Proceedings of 25th annual ACM symposium
on user interface software and technology, UIST’12, pp 589–598
17. Ishiguro Y, Poupyrev I (2014) 3D printed interactive speakers. In: Proceedings of 32nd
annual ACM conference on human factors in computing systems, CHI’14, Toronto, pp
1733–1742
18. Jones RO, Iravani P, Bowyer A (2012) Rapid manufacturing of functional engineering com-
ponents. http://opus.bath.ac.uk/30281/1/RapidManufacturingFunctionalComponents.pdf.
Accessed 28 Dec 2015
19. Vaezi M, Chianrabutra S, Mellor B, Yang S (2013) Multiple material additive manufactur-
ing—Part 1: a review. Virtual Phys Prototyping 8(1):19–50
20. Meisel NA, Elliott AM, Williams CB (2014) A procedure for creating actuated joints
via embedding shape memory alloys in PolyJet 3D printing. J Intel Mater Syst Struct
1045389X14544144
21. Malone E, Lipson H (2008) Multi-material freeform fabrication of active systems. In: 9th
ASME biennial conference on engineering systems design and analysis, pp 345–353
22. Paulsen J, Renn M, Christenson K, Plourde R (2012) Printing conformal electronics on 3D
structures with Aerosol Jet technology. In: IEEE future of instrumentation international
workshop, pp 1–4
23. Voxel8 (2015) Voxel8 3D electronics printer. www.voxel8.co. Accessed 28 Dec 2015
24. Espalin D, Muse DW, MacDonald E, Wicker RB (2014) 3D Printing multifunctionality:
structures with electronics. Intl J Adv Manuf Technol 72(5–8):963–997
25. Malone E, Rasa K, Cohen D, Isaacson T, Lashley H, Lipson H (2004) Freeform fabrication
of zinc-air batteries and electromechanical assemblies. Rapid Prototyping J 10(1):58–69
26. Toutonghi M (2015) Functionalize conductive filament blog. http://functionalize.com/blog/.
27. Anderson C (2006) The long tail: why the future of business is selling less of more, Hyperion
28. Sher D (2015) Google’s project ara looking at 3D printing for future. www.3dprintingindustr
y.com/. Accessed 28 Dec 2015
29. Meinders E (2011) Fast and flexible production. www.tno.nl/media/…/lr-leaflet-fast-and-
flexible-production21.pdf. Accessed 28 Dec 2015
30. Jones R, Haufe P, Sells E, Iravani P, Olliver V, Palmer C, Bowyer A (2011) RepRap—the rep-
licating rapid prototyper. Robotica 29(1):177–191
Chapter 10
From Mechatronic Systems to
Cyber-Physical Systems: Demands
for a New Design Methodology?
Peter Hehenberger, Thomas J. Howard and Jonas Torry-Smith
10.1 From Mechatronic Systems to Cyber-Physical

Systems
Mechatronics may be defined as an interdisciplinary field of engineering science

which aims to mutually integrate and interconnect mechanical engineering, elec-
trical engineering/electronics and computer science (also often called information
technology) such that the interactions constitute the basis for the design of suc-
cessful products [1–3].
In a mechatronic design process the conceptual design phase is especially
important. Here, the functional interactions between discipline-specific sub-
systems are determined, and therefore have to be carefully considered. This
implies that during the phases of conceptual and preliminary design, the designer
should be able to quickly and accurately evaluate those system properties that
result from design changes in the mechanical components as well as in other
components.
The successful development of complex mechatronic systems is only possi-
ble by close cooperation between specialists in the different disciplines involved.
Thus the design activities take place in a multidisciplinary environment, which
often involves engineers and other experts with different backgrounds [4]. In
order to enhance the performance of new products, the positive interaction
between different fields of mechatronics is increasingly being used. This may
result in an increased complexity of the product because the mere combination
P. Hehenberger (*)
Johannes Kepler University Linz, Linz, Austria
T.J. Howard
Technical University of Denmark, Kongens Lyngby, Denmark
J. Torry-Smith
Novo Nordisk, Bagsværd, Denmark

DOI 10.1007/978-3-319-32156-1_10
148 P. Hehenberger et al.
of discipline-oriented partial solutions will usually not provide the optimal result
attainable with an integrated system [5].
Mechatronic design emphasizes the integration between engineers skilled in
specific domains such as mechanics, electronics and software. The interactions
between product developers from these different disciplines are, however, often
hindered by insufficient understanding between the disciplines, and by missing
common platforms for modelling of complex systems [6]. As many sub-systems
are sourced from external suppliers, there is a need for both horizontal integration
within organizations and for vertical integration between the sub-system suppli-
ers and the suppliers of the full systems. Lee [7] defines Cyber-Physical Systems
(CPS) as the:
…. integration of computation and physical processes. Embedded computers and net-
works monitor and control the physical processes, usually with feedback loops where
physical processes affect computations and vice versa.
Such systems can be found in areas such as aerospace, automotive, energy,

healthcare, manufacturing, entertainment and consumer appliances.
For easy validation of requirements during the overall product development
process methods are needed for the decomposition of high-level system require-
ments into criteria for design decisions. This will be an important aspect in the
future, where new system architectures such as those of CPS are introduced [8, 9].
System modelling and evaluation are also important topics of CPS, for which
improved tools and knowledge are always required by the engineering profes-
sion. In many cases, very accurate system modelling is not a reasonable approach
to describing a complex system as the uncertainties and costs of even relatively
detailed modelling may be so high that the drawbacks compared to simpler model-
ling become overwhelming, so there is increasing trend towards system-level mod-
els which allow a multidisciplinary engineering approach to be supported.
CPS system-level models need specific methods, languages and tools to support
multi-view modelling in order to facilitate an interdisciplinary approach. More
generally, this objective can be realized through multiagent modelling, based on
an engineering cloud structure. This also results in the usage of tools supported by
Model-Based Systems Engineering (MBSE) [10].
CPSs are often dominated by one engineering discipline. System-level models
have to promote an equal treatment of all engineering disciplines involved during
product development and project execution.
Figure 10.1 presents a structure model describing the relationships between
mechatronic systems and CPS (shades of grey). These core elements are all com-
prised of a hardware element and a software element. The interactions between
these can occur in the physical domain (e.g. clash between two robots detected,
thanks to their sensors), or in the cyber part (e.g. dialogue between these robots
supported by network protocols). The cyber part is then considered as the inte-
gration network. All these core elements are themselves made up of several mod-
ules represented by the small white boxes. The mechatronic systems and the CPS
located on the border are then parts of the Systems of Systems.
10 From Mechatronic Systems to Cyber-Physical Systems … 149
Fig. 10.1 Mechatronics and cyber-physical systems as a systems of systems [6]
10.2 System Design Methods
In this section the issue of system design methods is presented considering the
hierarchy of systems and the used models.
10.2.1 Hierarchical Modelling
One of the major challenges in developing mechatronic products is the increasing

function density and complexity of such products as well as of the corresponding
mechatronic design processes.
The defining feature of mechatronic products, also known as systems of sys-
tems, integrated systems, or mixed systems, is that they merge solutions from dis-
parate disciplines. As a consequence, a mechatronic design process must integrate
multiple disciplines. There is a significant lack of such integration in traditional
design processes which in general have emerged more or less from a single disci-
pline and have incorporate other disciplines later.
The mechatronic design approaches (e.g. the V-Model of VDI 2206) focus on
exactly this integration aspect of the synthesis, analysis and evaluation steps in
modern mechatronic design. However, there are still a lot of open questions.
Hierarchical modelling concepts, i.e. the use of models with different granu-
larity for different levels of abstraction, are a promising approach to model and
master complex systems. The chosen approach is based on a modular structure of
models (modular model architecture, model base, hierarchical structure of models)
that allows for the configuration of system models from a library of sub-models
and interface models. A sub-model can be a model of a single mechanical com-
ponent or of a complex system, e.g. a model of an integrated mechatronic sys-
tem including an embedded control system. An interface model represents the
(physical) interaction between two sub-models, between sub-models and the sys-
tem model or between the system model and (the model of) the environment of
the system under consideration. This results in the advancement of reduced order
modelling as a key for coping with models of complex systems, the improvement
of the system view by system modelling and a significant advancement of the
mechatronic design process itself by more systematic approaches with particular
focus on the early phases of design (conceptual design, preliminary design).
The structure of the model of a CPS aims to store the product information
from the entire product life cycle. This requires a hierarchical structure, which
can be used in the early phases, which does not change during the product devel-
opment process. Normally, product information is structured according to either
the geometry or the assembly structure of the product. This results in problems
in the early phases of the product life cycle as the geometry or assembly structure
results from the development process and neither exists in the early phases nor
is stable during the development process. The object which is most stable dur-
ing product development is the set of resulting properties which define the prod-
uct. As previously stated, requirements can change over time, but to be precise,
changed requirements result in the definition and development of a new product.
Trying to find a product information structure which is suitable for every set of
requirements ultimately results in a structure which is identical for every product
development process [11].
On the other hand, we have network of interactions and properties influenced
by the different system elements. Basically, one type of property, a definable prop-
erty, can be any property the designer defines directly (e.g. materials, manufactur-
ing parameters, geometry). The totality of all definable properties then defines the
complete product with all its properties and its behaviour.
The resulting properties are used to structure the generally high number of
definable properties. This is done by assigning each definable property to the
resulting properties influenced by this definable property. As previously men-
tioned, it is possible that a definable property influences more than one resulting
property. For example, the definable property “material” influences the resulting
properties “maximum weight” and “maximum stress”). A second level of structur-
ing is achieved by assigning the definable properties to the different views. Each
definable property can appear in a single view or in multiple views. For instance,
the definable property “material” appears in the views “producibility” and “costs”.
In order to classify the definable properties in the matrix, a meta-information can
be assigned to each of them. In most cases more than one model can be used for
the description of a specific resulting property, whereas the less complex sub-
sidiary models were used in the early phases of the product development pro-
cess, when there is only little information about the product. During the product
development process more and more complex subsidiary models can be used, to
describe the interrelation between definable properties and resulting properties.
The knowledge about the interrelation between definable properties and result-
ing properties, gathered together from the subsidiary models can be used to set the
values of the definable properties, such as geometric dimensions, material type,
properties of the production processes. For example, the product designer uses a
FEM model to determine stress.
10.2.2 System Model
During all phases of the design process there is a need to build models which
may be seen as simplified representations of an original. In different phases these
design models have different goals. During the conceptual design phase, physical
principles, functions, structures, etc., have to be evaluated by executing models. In
the context of mechatronic system design processes, the phenomena under consid-
eration usually are of a physical or chemical nature. The models consist of a set of
parameters as well as a set of logical and quantitative relationships between these
models [4, 12].
The number of parameters increases from the conceptual design stage to pre-
liminary and detailed design. According to the increasing degree of detailing dur-
ing the design process, the granularity of the describing models becomes finer and
finer, leading to a hierarchy of models as well as their describing parameters. For
modelling and evaluation of solutions during all phases of the design process, we
postulate models with different degrees of detailing (granularity of models) in cor-
respondence with their describing parameters. The correspondence between mod-
els and parameters implies that the meaning of a parameter is well enough defined
via its related model. Hierarchical models are very important tools for complex
activities such as engineering design. Especially during the conceptual design
phase where there is a high demand for models to describe the design concept with
respect to the given requirements.
In addition to the hierarchical differentiation, the extensive functionality and
complex structure of mechatronic systems has the result that it is generally not
enough to optimize on a single criterion and a multi-objective optimization is
often needed. Optimization separately within each domain will not result in the
optimum system design. Therefore all the domains of, for instance, an automotive
sub-system have to be treated concurrently, at least in the beginning of the design
process. That way it will be possible to translate aspects from say the control or
electrical design to the mechanical design at an early stage of the design process.
This approach makes it possible to find a promising concept for the entire sub-
system and not only for a specific domain within the sub-system.
In order to master the mechatronic design approach and to benefit from it as

much as possible, a hierarchical design process is proposed in [11], in which
the discipline-specific design tasks need not be integrated as a whole on the
mechatronic level of the design task. Consequently, the system models should
cover the different views on a system as well as the different degrees of detailing
which lead from a hierarchy of models to a hierarchy of design parameters. Their
description by significant quantities is used for analysing different mechatronic
design concepts. The general requirements can be summarized as:
• Description of product information from all phases of the product life cycle tak-
ing into account the lifespan of a product as a necessity when storing the prod-
uct relevant information within a single structure.
• Coupled analysis of different physical product properties as caused by the
increasing integration of different applications and functionality in technical
systems, the demand arises to find a uniform description of the various product
properties, to define, describe, dimension, and calculate the product.
• Taking into account the view on an application domain as different development
phases and mechatronic disciplines require different views on the object to be
analysed. A general description of the requirement entails that model parame-
ters should be provided in various combinations and different granularity for a
further (more detailed) processing.
A major characteristic of mechatronic systems is that their properties are to a con-
siderable extent defined by software elements. This initially leads to a shift from
the physical function of the basic system to a realization through electronics or
information technology. As a consequence, with some adaptations, intelligent
systems of the future could include learning and decision-making abilities. Such
systems may include, for example self-optimizing processes to adapt to changing
conditions.
Crucial to the success of such a product is the behaviour of the overall system,
as customer requirements and desires almost always relate to the whole system
and not to partial sub-systems, components or even individual components. To
assess the characteristics of the system it is appropriate to use models. For mod-
elling and description of a mechatronic system, it is necessary to decompose the
system into a selection of suitable subdomains, describing the boundaries of the
considered mechatronic system to its system environment, enabling the flow of
matter, energy and information.
The functioning of mechatronic systems is distinguished from other systems
and hence requires the clear definition of interfaces and areas of responsibility.
It must therefore be clarified in what way interactions with the system environ-
ment (e.g. chemical, energy, information) have to take place. In the ideal case,
the whole system has the form of a cross-domain model. Unfortunately, the dif-
ferent disciplines use different modelling approaches and model descriptions.
Moreover, within the disciplines, information and data at high detail level are only
partially needed across the other disciplines. It is necessary to collect data for the
overall behaviour for understanding of the overall relevance. The objective is to
Fig. 10.2 System-level models [21]
create a system model with the information depicting the various domains, with
focus on the importance for other domains. The challenge is that the knowledge
of the entire system does not equal the sum of knowledge from the corresponding
domains. The domain knowledge must therefore be generalized (abstracted) and
integrated (see Fig. 10.2).
System-level models should at least facilitate management of existing data and
visualization of both the relationships inside a system (between its sub-systems)
and between a system and its environment. Additionally, they should provide the
possibility to execute several simulations of load cases (test cases), thus allowing
specific system properties to be evaluated. The simulations at the system level dif-
fer from those at the discipline level. Since simulations at the discipline level are
usually conducted by highly skilled and specialized engineers who use particular,
discipline-specific software tools, the simulations at the discipline level normally
cannot be replaced by simulations at the system level. Therefore, methods are
necessary:
• For modelling and simulation with special emphasis on the system view of the
design object, i.e. the object under design.
• For conceptual and preliminary design relying on concept models at an appro-
priate system level representing the essential information including a significant
system view of the design object.
• For the definition of modular structure of models (modular model archi-
tecture, model base, hierarchical structure of models) that allows for the
configuration of system models from a library of sub-models and interface
models.
• For the decision as to which information (e.g. which properties and which of
their quantifying parameters) should be included in the system model and which
in the sub-models.
10.3 Robust Design of Mechatronic Products and Systems
In this section the issue of robust design of mechatronic products is looked at,
and several challenges are raised with respect to the differences in perspectives
of the mechanical and electrical disciplines when it comes to reliability and
robustness.
10.3.1 Motivation for Robust Design
Robustness relates to the ability of a product to perform despite unwanted vari-

ation to parameters and noise factors. Robust design is therefore an incredibly
important methodology in mechanical design as it has the potential to allow manu-
facturing and assembly to have much looser tolerances, driving down the cost of
production and increasing the quality of the product.
The benefits of having a robust design are not just limited to the relaxing of
tolerances. In the news headlines and magazine reports, it is common to see sto-
ries of product recalls, high customer complaint rates, reduced product reliability,
unscheduled maintenance and repair and even major or catastrophic product fail-
ure, all being damaging symptoms of a lack of product robustness.
However, this is the tip of the Iceberg! The reliability and quality issues that
occur within a company due to a lack of robustness are large and costly. They
include delayed product launch dates, misplaced RD resources, increased quality
control and inspection and reduced innovation height.
While each discipline has its own approach to handling robustness, it is now
very common for the effects of variation to run from one discipline through to the
next, requiring robust design and optimization across disciplines, most typically
from the mechanical to the electrical domains.
10.3.2 Robustness as Hidden Reliability
It is clear from interacting with the reliability engineering community, that the
electrical engineers have a different tool sets and approaches to those adopted by
mechanical engineers with regards to product reliability. However, there are some
clearly overlapping methods.
Fig. 10.3 Occurrence of
failure modes [13]
For instance, Highly Accelerated Life Testing (HALT) is a methodology that

can test the reliability of both the mechanical and electrical aspects of a design
both independently and integrated into a system. HALT is used to test a single
product (perhaps of nominal dimension) often in its intended use conditions, but
with a highly accelerated use cycle. If we look at the nature of product failures in
industry, HALT will only give part of the story. Figure 10.3 [13] is an illustrative
example of the occurrence of the different failure modes. The failure modes with
high occurrence (labelled design failures) are easily detectable through product
testing such as HALT. However, the long tail of failure modes may occur perhaps
only one in 100 or one in 1000 and may not be detectable during testing. These
types of failures are termed robustness failures.
In addition to the above, it is clear that electronics reliability testing is domi-
nated by Weibull analysis and mean time to failure. This is dominated by the more
binary nature of electronics—it is working or it is not. Weibull would help to make
a probabilistic failure analysis of components and therefore with a fault tree analy-
sis (or similar) to calculate the probability of system failure at a given time.
In mechanical design, approaches such as kinematic design, design clarity
[14], location scheme design [15], axiomatic design [16] are typical approaches
to dampen issues when the product is not nominal (variation is present). The focus
here is not to predict product life but to predict and decrease the change in perfor-
mance due to variation. In an additional publication [17], 15 strategies for varia-
tion reduction are laid out. However, only two of these strategies (Quality Control
and Shielding) are possible in the electronics industry.
10.3.3 Electrical Mechanical Interfaces
All electrical components have a mechanical interface to them. Mounting and fit-
ting of PCBs is notorious for throwing up late stage issues and failures especially
during the high volume ramp-up. There are two common errors here. First it seems
to be a common mistake to overconstrain the PCB within the housing. An example
Fig. 10.4 Effect of overconstraint of PCB and housing. a Original design. b Redesign
of this can be seen in Fig. 10.4a where it is clear that there are too many surfaces
responsible for the positioning of the PCB within the housing (the outer edge, the
four pins, the inner edge, the notch at the top and the bottom). As a consequence
the PCB had to be reworked in various areas before the assembly process—the
areas of rework can be seen in Fig. 10.4b.
In the case above where a PCB is unable to fit correctly in its housing, the
worst case will be time and money spent on rework before assembly. In some
cases it may also lead to a potential distortion of the PCB during mounting, caus-
ing it some damage. When it comes to mounting of sensors, the positioning of the
sensor can have a huge consequence on the performance of the product, this is a
case of cross-domain robustness, dealt with in the next section.
10.3.4 Cross-Domain Robustness
With reference to mechatronics, cross-domain robustness arises where a variation

in the mechanical domain leads to performance change in the electrical domain,
or variation in the electrical domain affects the mechanical performance. There
are many examples of this in various signal processing applications. One typi-
cal example can be seen in Fig. 10.5 where a series of teeth are arranged to rotate
through a detector. The counting of the teeth by the detector determines both the
speed of rotation and the position of the rotating part.
Figure 10.5b shows how the teeth are returned from manufacturing where the
flash from the injection moulding introduces unwanted dimensional variation on
the teeth, which means that the signal generated was less precise and may even
reach a failure mode where a tooth is not recognized, producing a step change in
the velocity reading.
The signal produced from the detector and the tooth arrangement in Fig. 10.5a
is shown in Fig. 10.6. The signal processing software will then process the
Fig. 10.5 Rotating speed and position detector. a Teeth and detector. b Detail of teeth
Raw Signal
1500 MovAvg
TickIndicator
Difference Filter
Potential difference
1000
500
-500
-1000
3400 3500 3600 3700 3800 3900 4000 4100 4200 4300
Relative sample number
Fig. 10.6 Signal produced by the sensor of Fig. 10.5
differences between the peaks and the troughs on the signal in order to calculate
when a tooth has passed the detector. However, at certain geometric variations and
at higher velocities the detector can fail to identify the gaps between the teeth.
The above is quite a straightforward case where it is possible to do some sim-
ple robustness optimization so that the correct teeth numbers and dimensions are
designed so that the detector works over a wide range of velocities. The robust
optimization can also be undertaken for the microvariation of tooth dimension and
detector positioning and even the influence of noise factors such as temperature
causing thermal expansion. However, it is much easier for blind spots to occur
between domains where is it unlikely that the mechanical engineer will fully
appreciate the consequence of geometric variation on the electrical domain, and
the electrical engineer will often be unaware of the potential sources of mechani-
cal variation and the decisions that induce it.
This lack of understand between the domains is significantly compounded
when the complexity scales up. Design changes and optimizations can occur in
each discipline with not full appreciation of the effect on the other. A very good
example can be seen in the following case.
10.3.5 Modelling Across Disciplines
While there have been many valuable attempts to model mechatronics from a
number of different authors, the design and modelling of mechatronics is still a
considerable challenge [18]. In product development, an incorrect product inte-
gration within a system is clearly seen in the media. For instance, in 2014 the
GM ignition switch (see Fig. 10.7) recall was headline news, with record break-
ing costs/damages to the company of around $1.2 billion involving the recall of
28 million vehicles. The basic failure mode meant that the switch was unable
to provide the torque to hold the key in the ON position while the vehicle was
running, and in some circumstances the switch would slip from the ON to the
(ACC)ESSORY position.
The failure to identify and remedy the issues in this case were quite systemic
and reported at length by Valukas [19]. First, it is important to point out that
this was a component provided to GM by a supplier and which had to fit on
multiple GM vehicles. Both the switch and the series of vehicles each have their
own electromechanical systems which must work in harmony. In the case of
the GM ignition switch the mechatronic modelling approaches simply did not
support the engineers adequately, leading to major consequences. The follow-
ing three pieces of evidence are taken from the Valukas report to support this
argument:
Fig. 10.7 Recalled GM ignition switch

1. None of the mechanical engineers working on the ignition switch were aware
that moving the switch from the ON to the ACC position would shut off the
airbag along with the power-assisted brakes and steering.
2. Due to electromechanical issues, the switch was sometimes unable to be oper-
ated in cold weather. This was then confused with the slipping issue.
3. Delphi had not achieved the required torque for the ignition switch. Given the
switch’s history of electrical failures, however, they were hesitant to make any
changes that might jeopardize the functionality of the switch’s architecture.
The first item mentioned in the report is about understanding the functionality of
the products across domains. Without proper modelling of the functionality, engi-
neers face the challenge of being able to foresee problems caused by decisions
taken in one domain affecting other domains. This issue is not only present for
engineers working on systems with many components such as cars, but is also
present in products with fewer components and less complexity. The reason being
that most companies are driving performance of their products in the pursuit of
being one step ahead of their competitors. Inherently, this will result in many
issues to be solved within each of the domains. Focusing on the problems lying
ahead within each domain tends to attract engineers’ attention to their own disci-
pline-specific issues and away from integration issues; the so-called ‘silo think-
ing’. A failed interface between two domains may, however, just as well lead to a
break down or degraded performance of a product.
The robustness of a mechatronic product can be affected in one of the three fol-
lowing ways:
1. Robustness issues within a single domain. For instance, feedback issues in
electronic circuits creating unstable performance.
2. Robustness issues in the functional interaction between domains. For example,
a mechanical strain gate creates an out of specification signal at one end of the
performance spectrum, which creates an incorrect signal error when processed
by the electronics.
3. Robustness issues caused by adverse effects. Adverse effects are unforeseen
events creating an effect in domain other than where the event appears. Thus,
heat from electronic components could cause a bearing to lock unintentionally
due to the thermal expansion of materials.
To increase robustness, each domain must be internally robust (intra-domain
robustness) as well as the product have interfacing functions between domains
which are equally robust (inter-domain robustness). Some products may have
a clear-cut interface between domains posing the opportune situation that the
development can be divided into two different tracks with very little interaction
(e.g. between mechanical and electrical development teams). The trend, however,
seems to go in the opposite direction; namely to create more integrated products
with many functional interfaces between mechanical, electronics and software
solutions. This seems to have been the trend over the last decades and one might
assume that this trend will continue in the future.
So to be able to create robust products in the future we need methods to access

the inter-domain robustness as it seems that intra-domain robustness is better
understood, modelled and comprehended, although not perfected. Tools to assess
the level of complexity of the functional interfaces between domains as well as
evaluating the vulnerability of a single functional interface and its effect on the
robustness of the overall system are lacking. Fault tree analysis is one option
but is far from adequate in assessing the robustness of a full-blown mechatronic
system.
This speaks to the argument that the future for mechatronics research will
respond to the modelling requirement to identify failure modes and the effects of
variation across disciplines. A comprehensive modelling approach would lead to
more effective design change management and design optimization.
10.3.6 The Importance of a Cybernetic Perspective

on Robust Design?
In summary, it is quite clear that robustness optimization processes within a sin-

gle domain, not only give an incomplete picture when optimizing a product, may
actually reduce the cross-domain robustness for cybernetic products. This is not a
minor consideration by any means, as the cross-domain robustness issues are seen
as many both critical and main functions in many modern products. Examples of
which are drone technology, self-driving cars, production robots or the explod-
ing number of cybernetic products classed within the Internet of Things, all of
which are characterized for a need of mechanical precision, timing and perfor-
mance based on sensor inputs, signal processing. It would seem that defining
cross-domain modelling approaches that can be used in conjunction with new and
existing robust design approaches, not only defines the forefront of robust design
research but also the quality and reliability of many emerging CPS.
10.4 Challenges for an Integrated Design Methodology

for CPS
Design of CPSs involves close examination and further development of design

methods, design processes, models and tools. The current trend in mechatron-
ics involves networked mechatronic systems, or cyber-physical systems (CPS).
Therefore product lifecycle management and product data exchange play an
important role in CPS design. In order to push the performance of CPS and the
related design process, it is necessary to increase the research on system model-
ling, thus significantly improving the system view. There are many challenges
for future research towards improving the efficiency and quality of CPS product
development. In terms of model-based methods and tools in early design phases
which are the themes of a Design Society Special Interest group [20] who empha-
size challenges related to the following topics:
• Methods to generate solution concepts for CPS.
• Methods for evaluation of solution concepts.
• Methods for early concept optimization.
• CPS modelling aspects in terms of model integration, modular modelling, and
model interfaces.
• Performance of CPSs (multivariable key performance indicators, methods for
cross-domain aggregation for performance evaluation).
• Methods and tools supporting CPS Design.
• MBSE for CPS.
• Integration of the cyber and the physical sub-systems.
• Cyber-physical robustness.
• Information and knowledge flow during CPS design processes.
• Modelling engineering change propagation and functional couplings in CPSs.
• Description and modelling languages for all design phases of CPS.
• Closing the gap between scientific approaches and industrial practice.
The key to an integrated design methodology for CPS is modelling and simula-
tion (see Fig. 10.8). In this context design models and their interchangeability
between different design tools are very important during the design process. From
the mechatronic design process viewpoint, models are containers of the knowledge
of the product during its total life cycle. Simulations are producing information of
the design problem. This may improve product knowledge and potentially also the
quality of many analyses and decisions. The presented approach relies on modular
model architecture and enables innovative design, flexibility, speed and assistance
in nonroutine design questions.
The main point of view of the work is “simulation”, namely modelling and vir-
tual experimentation regarding the behaviour of a system. The considered objec-
tives of the simulation are:
• The creation of executable models of design concepts at the system level and
the evaluation of alternative concepts.
• The creation and production of multidisciplinary simulation models at the sys-
tem level of CPS.
• Simulation models for the use in system testing.
Linked to these objectives is the question of the required (mathematical) descrip-
tion of different types of simulation models. In this context, the aspect of usage
and reuse of (simulation) models is an important topic. There is a lack of meth-
ods and software tools supporting the modelling and simulation aspects in the
early phases of product development, during which detailed models due to
incomplete information cannot be established and therefore the system has to
be modelled on a level with high abstraction. It is a challenge for the future to
derive the requirements for such tools and to develop appropriate software for
this purpose.
Fig. 10.8 Challenges for the future
For a simulation-based engineering approach, a model-based description of

the system, with respect to its sub-systems, under consideration is a prerequisite.
Especially for mechatronic systems this addresses:
• The models of the sub-systems.
• The integration of these models to an overall CPS—system model.
This engineering approach facilitates a holistic view on the overall system and
should be continuously applied even from the very beginning of product develop-
ment. Thus it becomes possible to model aspects such as requirements, functions,
behaviour and structure in an integrated way which is essential especially during
the early phases of design.
Acknowledgements We would like to thank all the members of the Design Society Design
CPS Special Interest Group [20], who actively contributed to the workshops and discussions.
References
1. Harashima F, Tomizuka M, Fukuda T (1996) Mechatronics—what is it, why, and how?

IEEE/ASME Trans Mechatron 1:1–4
2. Tomizuka M (2000) Mechatronics: from the 20th to 21st century. Control Eng Pract
10(8):877–886
3. VDI Guideline 2206 (2003) Design methodology for mechatronic systems @ www.vdi.eu/
guidelines/vdi_2206-entwicklungsmethodik_fuer_mechatronische_systeme/. Accessed 28 Dec
2015
4. Hehenberger P, Poltschak F, Zeman K, Amrhein W (2010) Hierarchical design models in the
mechatronic product development process of synchronous machines. Mechatronics 20:864–875
5. Isermann R (2005) Mechatronic systems: fundamentals. Springer, Berlin
6. Hehenberger P, Bricogne M, Le Duigou J, Eynard B (2015) Meta-model of PLM for Design
of Systems of Systems. Proceedings of the PLM international conference, PLM2015. Doha
7. Lee EA (2008) Cyber physical systems: design challenges. In 11th IEEE international con-
ference object oriented real-time distributed computing, ISORC, 363–369
8. Ordinez L, Alimenti O, Rinland E, Gómez M, Marchetti J (2013) Modelling and specifying
requirements for cyber-physical systems. IEEE Latin Am Transac 11(1):625–632
9. Simko G, Lindecker D, Levendovszky T, Jackson EK, Neema S, Sztipanovits J (2013) A
framework for unambiguous and extensible specification of DSMLs for cyber-physical sys-
tems. In 20th annual IEEE international conference and workshops engineering of computer
based systems, ECBS, pp 30–39
10. Haveman SP, Bonnema GM (2013) Requirements for high level models supporting design
space exploration in model-based systems engineering. Procedia Comput Sci 16:293–302
11. Hehenberger P (2012) Advances in model-based mechatronic design. Trauner, Verlag
12. Follmer M, Hehenberger P, Punz S, Rosen R, Zeman K (2011) Approach for the creation
of mechatronic system models,. In: Proceedings of the international conference engineering
design, ICED 2011:15.-19.08
13. Krogstie L, Ebro M, Howard TJ (2014) How to implement and apply robust design: insights
from industrial practice. Total Qual Manage Bus Excellence (ahead-of-print):1–19
14. Christensen ME, Howard TJ, Rasmussen JJ (2012) The foundation for robust design:
enabling robustness through kinematic design and design clarity. In: 12th International
Conference Design, pp 817–826
15. Söderberg R, Lindkvist L, Carlson JS (2006) Managing physical dependencies through loca-
tion system design. J. Eng Des 17(4):325–346
16. Suh NP (2001) Axiomatic design: advances and applications. Oxford Series on Advanced
Manufacturing, USA
17. Ebro M, Howard TJ (2015) Robust design principles for reducing variation in functional per-
formance. J Eng Desl (In Press)
18. Torry-Smith JM, Qamar A, Achiche S, Wikander J, Mortensen NH, During C (2013)
Challenges in designing mechatronic systems. J Mech Des 135(1):011005
19. Valukas AR (2014) Report to board of directors of general motors company regarding igni-
tion switch recalls, general motors internal investigative report. Jenner & Block, USA
20. Design society sig methodology for design, integration, modelling and simulation of cyber
physical systems (Design CPS) @ www.designsociety.org/ds_group/30/methodology_for_
design_integration_modeling_and_simulation_of_cyber_physical_systems_design_cps.
21. ACCM Multi-Firm-Project, System Models for Mechatronic Design, SyMMDe (2013–2016)
@ symmde.jku.at/. Accessed 28 Dec 2015
Chapter 11
The Internet of Things: Promise
of a Better Connected World
George R.S. Weir
11.1 Introduction
In recent years, progress in network-based applications has allowed a move

beyond the staple asynchronous communication of email in which each party
takes turns to compose and send a digital variation on traditional postal mail.
Increasingly, synchronous (live) interaction between parties is possible through
audio, video and typed digital channels by means of applications such as Skype,
Facebook Messenger and Apple Facetime.
Although email remains a prevalent medium, users have also come to embrace
social networking as a basis for selective broadcast and group interaction. These
innovative applications are widely adopted across sectors, age groups and nations,
in the take-up of laptops, smartphones and tablets. Of course, the increased use
of networked devices reflects an associated growth in networking infrastructure.
Wireless communication is normal practice in the use of mobile devices as well as
a convenient basis for local area networks.
Increasing Internet use and the ready availability of connected information
is often regarded as a natural step toward a greater degree of interconnectivity
in which many of the devices in our homes, offices and factories become linked
and capable of communication and control via local networks. In this chapter,
we consider how the current context gives rise to the ideas behind the Internet of
Things (IoT), look at how such extensive systems would function, and consider
what benefits and disadvantages we may expect from such developments. As well
as considering the present state of play, we will review the key ingredients, likely
applications, ecosystem requirements, potential issues and prospects for a happy
future enabled by IoT.
G.R.S. Weir (*)

University of Strathclyde, Glasgow, UK

DOI 10.1007/978-3-319-32156-1_11
166 G.R.S. Weir
11.2 Impetus
There are several factors in our current technological context that naturally direct
developments toward the extended integration and enhanced data exchange
that is core to the Internet of Things. On the one hand, there is familiarity with
increasingly functional and immediate communication facilities, with the asso-
ciated expectation that other information systems will be equally immediate and
responsive. On the other hand, rising service costs are an impetus toward wider
deployment of networked devices, since such developments are seen as a means to
cheaper service provision and, especially, service monitoring. Thus, there is grow-
ing anticipation of integrated systems that afford greater convenience, new services
and more economical provision of existing services. We may assume that ‘A typi-
cal home will soon contain a network of gadgets designed to make life easier’ [1].
11.2.1 Government Initiatives
In the UK, a report entitled ‘The Internet of Things: making the most of the Second
Digital Revolution’ [2] was prepared by the UK Government Chief Scientific
Adviser. In the USA:
… the Federal government is working now to direct development and testing of such sys-
tems with an eye toward a variety of future applications. The US government calls such
technology “Cyber-Physical Systems” (CPS) and is looking for ways they can be used to
improve safety, sustainability, efficiency, mobility and the overall quality of life [3].
In a similar vein:
The Singapore government has introduced a slew of initiatives as part of its goal to
become the world’s first smart nation, including a smart nation operating system, Internet
of Things scheme targeted at homes, and pilot trials at a designated residential-business
estate. [4]
Nations with developing economies are also rising to the IoT opportunity. For
instance in India:
One of the top most initiatives in the form of Digital India Program of the Government
which aims at ‘transforming India into digital empowered society and knowledge econ-
omy’, is expected to provide the required impetus for development of the IoT industry
ecosystem in the country” [5]
Each of these national perspectives reflects the view that engaging with IoT
developments will enhance the welfare of the population and the economic ben-
efit of the country. What then are the required ingredients for such progress in any
nation?
11 The Internet of Things: Promise of a Better Connected World 167
11.2.2 Key Ingredients
The UK government report [2] identifies three key ingredients in the Internet of
Things ecosystem:
(i) Communication;
(ii) Integration;
(iii) Data analysis.
First among these ingredients is the present and evolving communication infra-
structure, comprising existing ‘fixed’ network facilities, in addition to wireless
technologies, such as Wireless LAN (WLAN), Bluetooth, GPRS (GSM) mobile
telephony standards and anticipated new standards for ‘near-field’ and close-prox-
imity device interaction.
Integration is considered essential since the scope for IoT will depend upon
the consolidation of diverse systems and standards, with ‘local’ systems talking to
each other and to ‘upper level’ systems. Finally, data analysis appears in two roles.
First, such analysis serves as a means of monitoring and managing the quality of
interaction between devices (e.g. for fault detection), and second, as a value-added
ingredient that provides insight on usage and performance. (e.g. for targeting
bandwidth and premium enhancements). The expectation is that integrated sys-
tems will support widely distributed data gathering as well as centralised synthesis
and analysis of data.
11.2.3 Applications
Within the UK government report, five core sectors are identified as having major
potential to boost the UK economy through IoT developments
(i) Home automation
(ii) Agriculture
(iii) Energy
(iv) Healthcare
(v) Transport
For each of these sectors, we can anticipate IoT applications with significant eco-
nomic potential. Home automation should have wide appeal and would apply not
only to individual dwellings but also in the context of larger-scale building man-
agement systems designed to coordinate multiple interior systems, such as air con-
ditioning, temperature and lighting. Small-scale automation facilities are already
available for home use. These include ‘smart thermostats’ that are network-acces-
sible for remote control. Production and yield management in agriculture and
other manufacturing contexts stand to benefit from the introduction of sensor-
based feedback and automation.
168 G.R.S. Weir
Fig. 11.1 Anticipated growth of in-car IoT applications (after Smart Cars and the IoT [10])
The energy sector has already shown movement in the direction of IoT through
introduction of smart metres. These systems go beyond mere recording of total
energy consumption to reporting consumption and usage patterns to the provider.
Healthcare is an important application sector for IoT primarily from a cost effi-
ciency perspective. The prospect of reduced cost health services through remote
delivery (eHealth) is an eagerly anticipated economic boon for a presently over-
stretched and cash-strapped National Health Service.
In the transport sector government advisers predict significant growth in the
use of in-car sensors, telemetry and inter-vehicle communication, as a basis for
self-driving vehicles (Fig. 11.1). Progress in such smart transport is illustrated by
the Cooperative ITS Corridor, an EU project to manage cars from Rotterdam via
Munich, Frankfurt and on to Vienna [6].
Roads equipped with cameras every 100 m and WiFi antennas every 500 m,
combine with short-range ‘car-to-road’ communication, in order to measure the
exact position of vehicles 10 times per second, within 1 m accuracy. Among the
perceived benefits are improved flow management, such as addressing the ‘brak-
ing shockwave’ problem on motorways, warning drivers of upcoming roadwork
and other obstacles. Such initiatives also aim to harmonize smart-road standards
among different countries. At first, such systems only employ ‘car-to-roadside’
communication, with plans to extend this later to ‘include car-to-car’ interaction.
While these anticipated economic benefits are central to IoT promotion by gov-
ernments, we can already see relevant devices and technologies entering the mar-
ketplace that will contribute to the adoption and growth of IoT. The prevalence
of home WiFi networks affords a convenient infrastructure for introducing the so-
called ‘smart’ devices with network communication capabilities. These vary from
domestic appliances such as toasters and kettles, through wirelessly controlled
light switches and multi-room digital music systems, to toothbrushes that report
the effectiveness of their use through Bluetooth connectivity. In the home context,
control facilities are readily afforded through mobile telephone apps or apps for
Android and iOS tablets. These examples illustrate the potential integration of
seemingly disparate systems.
In the realm of mobile systems, smartphones already support WiFi, Bluetooth

and near-field connectivity. In conjunction with in-built GPS capability and suit-
able software applications, these phones can seamlessly interact with the local
environment, registering their presence (or the presence of the telephone user),
registering relevant localized data for presentation to the user and reacting to per-
sonalized settings or user preferences, based upon time of day and geographical
position. Increasingly appearing as supplements to the ever-present smartphone,
we find smart watches, fitness trackers and other wearable devices, such as clothes
with in-built sensors. In keeping with domestic device development, these wear-
able devices build upon the functionality embedded in telephones and tablets to
engage data processing and communication facilities. A case in point is the wear-
able PoloTech™ smart shirt from Ralph Lauren that measures the wearer’s heart
rate and respiration, distance travelled and calories burned, with data transferred to
smartphone or tablet via Bluetooth.
11.2.4 Ecosystem Requirements
To appreciate the variety of prospective applications, we should consider the range

of device types, the networking modalities and the methods of communication that
are likely to comprise the essential infrastructure or ecosystem for the Internet of
Things. One essential aspect of such technologies is the inherent flexibility that
arises from multiple scales of device (with differing capabilities), different means
of establishing intercommunication with other devices and a variety of alternative
network topologies to suit differing needs. In terms of device capability, and asso-
ciated scale, we may distinguish three device varieties, characterised by the roles
that they play:
(i) Location markers (Passive).
(ii) Data gathering and relay (Active).
(iii) Decision-making (Active).
While we may naturally think of computation and data processing as necessary
features of IoT devices, considerable utility can be added through the use of ‘pas-
sive’ objects as components within a local network. Such objects are passive in the
sense that they have no native facility for generating, sensing or processing data.
Instead, they are able to signify their presence through use of ‘location markers’.
These markers may be based on Radio Frequency Identification (RFID) tags that
can be detected by RFID sensors. Such ‘smart labels’ may be battery-powered and
actively send their ID by radio waves or simply wait to be read by an active RFID
reader. The sole purpose of such tags is to signify the identity and presence (loca-
tion) of the objects to which they are attached. The objects and the tags may be
passive but detectable by other active systems. This allows for detecting or track-
ing tagged items and transfer of such information to local or remote computers.
170 G.R.S. Weir
The second variety of device has the native facility to capture and relay data.
This requires some sensor capability but, while in this sense active may have little
or no data processing capacity. The principal role for these sensor-based devices
is to gather local data and relay this to other more sophisticated devices where the
data from multiple sensor devices will be collated, aggregated and, perhaps, ana-
lysed. Our third variety of device covers those that actively process received data.
This includes any active device that receives sensor information directly or indi-
rectly, via other sensor systems. Combinations of these three device varieties sup-
port a hierarchical structure that allows data to be passed ‘upstream’ from multiple
sources to be collated and analysed; potentially, from local through district and
regional to national and beyond.
This hierarchy of interlinked components will rely upon several types of net-
work topologies. There will be scope for close-proximity communication based
upon an ad hoc network topology. This will support interaction from device to
device in cases if these devices are at a similar level of data gathering and distribu-
tion (i.e. peer to peer). From sensor-based, passive items and mobile devices, data
will be communicated to local networks and is likely to rely upon current tech-
nologies, such as WLAN and Bluetooth. In turn, LANs have connection through
Internet Service Providers to wider area networks. Thereby, the different network-
ing models will integrate and interact to provide an infrastructure at different lev-
els of complexity.
A complementary perspective on these networking models considers the inter-
actions between devices and systems in terms of communication. These models
represent the mode of interaction between different devices in the networking
context. A common interaction model is client–server (Fig. 11.2). This is the tra-
ditional form of Internet communication in which many smaller-scale systems
interact via one or more larger-scale systems.
Another ‘style’ of communication between systems that has become common
on the Internet is peer-to-peer interaction. This is characterised by systems or
devices communicating directly with other similarly scaled systems (Fig. 11.3).
A less common approach to communication is also feasible. In this case, indi-
vidual devices communicate with a central system that provides a repository of
data and results. This allows each device both to deposit and to query the central
Fig. 11.2 Client–server
communication model
Fig. 11.3 Peer-to-peer
communication model
Fig. 11.4 Blackboard
communication model
repository. This may be described as a ‘blackboard’ model (Fig. 11.4) and shares
many features of what has come to be called cloud computing.
The likelihood is that any significant installation associated with IoT would
engage several of these communication models, while most individual compo-
nents may employ a single model, most probably, the client–server or peer-to-peer
approach.
With the increasing presence of communications infrastructure and interoper-
ability of mobile devices comes new possibilities in tracking and monitoring of
domestic objects outside the home—children, pets, vehicles, mobile phones and
people. Of course, this is a double-edged sword that promises utility but also raises
issues of civil liberty and personal privacy.
As part of a domestic or commercial IoT ecosystem, we have the promise of
smart inventory, regulated service reports and associated ease of auditing and data
production (e.g. for insurance purposes or home reports when selling property).
Other features in prospect are highly integrated monitoring and control of heating,
cooling and energy management at the domestic, district, regional and national
level. Such environment monitoring for smart control may embrace ambient fea-
tures and anticipated changes, e.g. weather forecast affecting thermostat settings.
172 G.R.S. Weir
With more devices becoming ‘smart’ and able to register their status with
upper-level systems, we should expect increases in device self-monitoring for fault
tolerance and timely repair, e.g. as we have currently for vehicle engine status
monitoring. Significant cost benefits may arise through better insight on system
demand and better understanding of system performance. Allied to this may be
quality of service enhancements through optimized device and system design as a
result of greater performance feedback.
The costly realm of healthcare may expect to benefit substantially from remote
diagnosis and treatment, as well as through operational enhancements. In the
short-term, we may look forward to a more accessible, efficient and cheaper health
service.
11.3 Issues and Challenges
Among the likely issues that are emerging or will emerge in consequence of wide-
scale adoption of Internet of Things are the following:
1. Usability
2. Reliability and robustness
3. Availability
4. Locus of control
5. Privacy
6. Integrity
7. Security
11.3.1 Usability
Inevitably, with developments in ubiquitous technology there will be many diffi-

culties that arise through inherent system complexity, or through system misuse.
Some of the issues associated with Internet of Things are already evident in the
nature and use of today’s infrastructure. Other issues are predicted according to
the manner and mechanisms that will sustain the growth in IoT.
From a user perspective, usability is always a major concern but key to pro-
gress in IoT is the trend toward ‘invisible integration’. As domestic and commer-
cial items gain connectivity and native ‘intelligence’, these facilities may become
inherent and unseen, with little or no requirement for user activation or direction.
In other words, the essential aspects of IoT may be invisible in their usual opera-
tion. If this is accomplished, and this may be more aspirational than realistic, then
IoT technology will add little to any usability issues with connected devices.
11.3.2 Reliability and Robustness
Serious concerns associated with the reliability and robustness of devices and sys-
tems that constitute the Internet of Things are bound to arise. With greater depend-
ence upon such integration comes greater risk. If complex integrated systems
become mission or life critical, we will require assurance of reliability. This may
require insight on minimum failure rates for critical devices and their higher-level
systems.
With increased complexity we have multiple points of failure. The robustness
and reliability factors affect individual devices, communication links, centralised
and de-centralised services. Reliability is determined not only by failure rates
or how robust are the constituent parts, but also in terms of capacity and associ-
ated levels of performance. Quality of service may be critical just as absence of
device failure may be critical too. As with present day Internet connectivity, when
demand increases, infrastructure capacity has a direct effect upon service perfor-
mance. If there is a need to assign priorities and manage contention, then some
services, and probably, some users, will lose out.
11.3.3 Availability
The issue of availability is closely allied to the concern for reliability and robust-
ness. If system capacity is limited or not entirely reliable, how do we spread the
benefits? Unless there is equal service provision (or at least, availability) for
all, we risk a new era of ‘haves’ and ‘have nots’, in which the privileged (or the
wealthy) have greater access, availability or performance than others. The prospect
of emerging social benefits from IoT may herald a new realm of inequality of ser-
vice availability determined by cost of provision or geographical location.
Perhaps we should expect differing service options at different costs. One
case in point may be the rise of a two-tier national health service with two access
modes: personal contact and online. Presumably, the latter will initially be the
cheaper option but this might evolve into a more specialized service, e.g. advice
and input from world leading medics, at a premium cost.
11.3.4 Locus of Control
Since IoT introduces major scope for data gathering and assimilation, the issue
of control will concern many individuals and organizations. Current data gather-
ing points, such as popular search engines, already raise questions of ownership,
control and use of information. Similar questions arise regarding state access and
use of information. If individuals yield control of information about their online
174 G.R.S. Weir
and offline behaviour, they lose influence on how such information may be used.
Optimistically, the information will be used positively to optimize services and
minimize costs. Pessimistically, there may be adverse effects upon particular indi-
viduals or organizations, such as members of groups that are perceived as radical
in their social, political or religious views.
In response, one might suppose that IoT leaves little scope for individual or
local control of information. Indeed, one may argue that any ‘added value’ arising
from the synthesis of data depends upon the aggregation of many data sources.
Nevertheless, by its nature, the envisaged data aggregation requires author-
ised access to data that is ultimately derived from individuals or the systems and
devices belonging to those individuals—and this naturally leads us to the issue of
privacy.
11.3.5 Privacy
In this envisaged context of centralised data collection, we presuppose the appli-

cation of data analytics across ‘big data’. As well as the aforementioned issue of
control, we may ask ‘Who owns the information?’ and ‘Who determines how it
may be used?’ Given that some people may wish to withhold information, can this
be accommodated within the wider system? If not, can we secure guarantees that
information in which we figure cannot be used in adverse effect against us? Along
side the prospective benefits of timely intervention, e.g. based upon an individu-
al’s biological data, comes threats to privacy and civil liberty, e.g. through ‘timely
intervention’ and removal of health insurance benefits based upon an individual’s
biological data. Likewise, freedom of movement may be devalued if individuals
are tracked via their use of mobile systems and have ‘nowhere to hide’.
On a less sinister note, collective data, e.g. associated with product perfor-
mance and use, may hold great value to device manufacturer but afford little or
no direct benefit to individual users. In the absence of incentive to contribute such
data, will individuals have scope to opt out? More likely, participation will become
a condition of system provision. If you want the service, you contribute the data.
Availability of data may raise questions over who will have access to such
information. Increased resources of amalgamated data may generate new scope for
data brokers and will certainly herald new avenues for personalized adverts.
11.3.6 Integrity
As we become increasingly dependent upon systems that relay information to

higher-level systems, for data integration and analysis, questions may arise in our
minds about the conclusions drawn from data that we have contributed. Assuming
that the results are actually available for our inspection, can we trust the results
of data analysis? Is there any scope for independent verification? At the domestic
level, as well as relaying data to the supplier, smart metres may provide consumer
feedback on energy usage. Access to the raw data and the basis for supplier cost
calculations should allow us to determine the correctness of any resultant charges.
But will all automated data transfer systems afford such transparency to the con-
sumer? Alternatively, will intermediaries, such as industry watchdogs, have a role
in policing the integrity and quality of such services?
11.3.7 Security
The security of systems and devices is our final area of concern with the Internet
of Things. The preponderance of devices will only be as strong as its weakest
point and we may expect many weak points in the explosion of interconnectivity
arising from IoT. In anticipation of this issue, some have even dubbed the develop-
ment ‘the Internet of Insecure Things’, with the depressing thought that ‘Anything
that can be hacked will be hacked’.
Evidence from existing networked systems and devices reinforces this unfor-
tunate prospect. For instance, malware (allegedly originating in China) has been
found on US SCADA (control) systems. Many nation states have growing anxiety
over risks to national infrastructure, as evidenced by examples of attacks on the
US power grid. A demonstration under Project Aurora, illustrated such vulnerabil-
ity to attack, with a $1 million diesel-electric generator destroyed as culmination
to the experiment. The frequency of data breaches is further indication that inter-
connections between systems may give rise to weaknesses as well as strengths.
One might suppose that developments in the form and function of newer
devices would include protection against such risks. Yet the vulnerabilities per-
sist primarily because the forms of attack are still effective. As previously noted,
increasingly complex systems have more potential points of failure. Any party
seeking mischief against an IoT installation may target individual devices or target
the network and communication infrastructure. Most attacks use standard proto-
cols to overwhelm the target. Since the connectivity and communication protocols
are fundamental aspects of the system, they cannot be disabled as a defence. In
consequence, any connected device will be vulnerable, by its nature. The inher-
ent risks are unauthorized access (to data or control). With network access to a
device, an intruder may retrieve stored data from the device or modify the device
behaviour by means of remote commands or re-programming the device’s stand-
ard behaviour.
Several prime examples of remote tampering have come to light recently. A
case in point is the Internet-enabled fridges that use email to communicate their
status [7]. In one instance, hackers have successfully gained access to the software
system in such fridges and changed the programming to send spam emails. Similar
remote access problems often affect Internet-enabled devices, including wireless-
linked cameras.
176 G.R.S. Weir
Recent press stories report Web sites offering lists of remote cameras that can
be viewed from anywhere on the Internet (without the permission, approval or
knowledge of the camera owners). In one example, a Web site was found to be
offering links to unsecured security cameras in 256 countries [8].
Remote access is often achieved by guessing a factory-set password that allows
user control of the facility. Commonly, such devices are installed without change
to their pre-set access passwords. Leaving them vulnerable to any remote user who
can locate the device on the Internet and determine the required authentication
details. The risk that unauthorised individuals may misdirect devices or acquire
personal data from associated systems is significant and a realistic concern. In
addition, experience shows that simpler remote interference with networked
devices can impair or deny the service to legitimate authorised users or disable the
normal operation of the device and its associated service.
Such interference is aptly termed ‘denial of service’ and attacks of this nature
often occur against Web services. In each instance, the attack is designed to fully
engage the system and, usually, disable it through overloading its network inputs.
Often, the technique will direct network traffic to the target service from many
other devices that have been compromised, taken over and controlled remotely,
without the knowledge of their owners. Such distributed denial of service attacks
may simply overwhelm the limited capacity of the target to handle incoming com-
munication or service requests. The assailed system may simply ‘crash’ and cease
to operate or fail to perform its normal operation while it is buffeted by the net-
work onslaught. Such attacks may result in service disruption, data loss and asso-
ciated damage to the public image of the affected organization.
The motivation behind such attacks may be mischief, political alignment or
extortion against the owner of the target system. In the IoT context, the risk from
denial of service attacks may range from inconvenience through financial loss and
public image impairment to physical injury or death. Especially in a setting where
we have implanted networked medical devices, the associated health risks from
illicit access may be severe. This risk is recognised in the decision by former US
Vice-President Dick Cheney, when undergoing heart surgery to have the wireless
connectivity disabled on the implanted defibrillator [9].
11.4 The Future Internet of Things
The Internet of Things is not a utopian ambition. Technology exists that will ena-
ble many of the applications described in this chapter, and many more to be speci-
fied as the vision expands. In addition, the lauded potential social and economic
benefits are plausible but not guaranteed.
As with many developments in technology, we may expect benefits and draw-
backs. On the positive side, there are clear indications that the extensive integrated
communication infrastructure that is fundamental to IoT will afford enhanced ser-
vices through wider automation, information access and exchange. Those users
who are able to quickly adapt and adopt the new technologies are most likely to
benefit from these developments.
On the negative side, for a variety of social, financial or educational reasons,
many prospective beneficiaries may be slow or ultimately unable to embrace the
new opportunities that arise from IoT. Alongside the social and economic bene-
fits, we may anticipate a new digital divide that arises from limited availability and
incompatibility of old and new technologies. This gap may be amplified if some
in society are unable to afford or to comprehend the technology and its potential,
while others remain relatively unmoved and disinterested. Some may be content to
adopt personal applications, such as health and fitness monitors or limited domes-
tic management systems. The majority may rush to join the advance. The signifi-
cant prospective impact of the Internet of Things lies in its broader application for
social change and economic transformation. Achieving this potential depends not
solely upon developments in technology but upon equitable access and affordable
opportunity.
References
1. Sunday Times (2015) 11 Jan 2015

2. www.gov.uk/government/uploads/system/uploads/attachment_data/file/409774/14-1230-in-
ternet-of-things-review.pdf. Accessed 31 May 2015
3. us.sourcesecurity.com/news/articles/co-5188-ga-sb.13927.html. Accessed 31 May 2015
4. www.zdnet.com/article/singapore-unveils-plan-in-push-to-become-smart-nation/. Accessed
31 May 2015
5. deity.gov.in/content/draft-internet-thingsiot-policy. Accessed 31 May 2015
6. c-its-korridor.de/?menuId=1&sp=en. Accessed 28 Dec 2015
7. www.theguardian.com/technology/2014/jan/21/fridge-spam-security-phishing-campaign.
8. www.networkworld.com/article/2844283/microsoft-subnet/peeping-into-73–000-unsecured-
security-cameras-thanks-to-default-passwords.html. Accessed 28 Dec 2015
9. www.bbc.co.uk/news/technology-24608435. Accessed 28 Dec 2015
10. www.abiresearch.com/market-research/product/1019236-smart-cars-and-the-iot/. Accessed
28 Dec 2015
Chapter 12
Home Technologies, Smart Systems
and eHealth
Jorge Azorin-Lopez, Andres Fuster-Guillo, Marcelo Saval-Calvo

and David Bradley
12.1 Introduction
The concept of a Smart Home, Digital Home or Domotics is based around the
deployment a range of technologies to provide features and functions related to the
management of the domestic environment [1]. Key components of such systems
are:
Sensors Provide information on the environment and its users

Actuators Provide and perform actions based on the interpretation of the sensor
data
Controller Analyses and interprets the sensor data in order to generate the appro-
priate actions in response
Smart devices Individual devices integrated within the network providing a range of
smart functions
Internal Integrates devices within the home network and provides link to
communications external communications as required
User interface Enables the user to interact with the system to define operating
parameters and set context as appropriate
Referring to Fig. 12.1, the general functionality of the home system can then be
considered in relation to:
J. Azorin-Lopez · A. Fuster-Guillo · M. Saval-Calvo

University of Alicante, Alicante, Spain
D. Bradley (*)

DOI 10.1007/978-3-319-32156-1_12
180 J. Azorin-Lopez et al.
Fig. 12.1 Functionality of a smart or digital home
Cloud Interaction with Cloud-based systems to provide a range of facili-

ties and services
Comfort This implies the operation of the environmental control systems to
provide optimum comfort range for individuals within the home,
including zoning to meet variations in individual user criteria and
remote access as required
External Supports the ability of individual systems within the home environ-
communications ment to access relevant, and context-based, information relative to
their operation
eHealth and mHealth Of increasing significance in the face of an ageing population and
implies the introduction of a range of monitoring and response
systems tailored to individual need
Home systems Smart homes will inevitable and increasingly contain a range of
smart, networked systems covering activities from entertainment to
domestic functions such as washing, cleaning and cooking
Security Implies both physical and cyber security under the general manage-
ment of the home systems
Smart grid Integrates individual homes within a group of homes and hence
with the energy supply grids to provide efficient energy manage-
ment and resource utilisation
12 Home Technologies, Smart Systems and eHealth 181
The implementation of smart home and related technologies involves a number

of systems issues for the short, medium and long terms such as the choice and future
proofing of technologies, ethics and costs. The chapter begins by looking at the back-
ground to, and structure of, smart home technologies and systems before progressing to
look at one particular area of application, that of eHealth and mHealth, in more detail.
12.2 From Domotics to Ambient Intelligence
Building automation services were initially provided by a set of non-integrated sub-

systems (heating, light control, fire alarm systems, escalators, etc.) within large build-
ings [2–5]. By the late twentieth century, this was starting to include home automation.
The introduction of the Internet then supported new concepts such as the
Digital Home, eHome or iHome [6–8] and saw the evolution of the traditional
automation services to include entertainment and communication supported by
home networks and residential gateways [9, 10].
The twenty-first century also brought new paradigms such as “ubiquitous com-
puting” [11] and “ambient intelligence” [12, 13], whose intent is to bring “intel-
ligence” to the environment. In the case of ambient intelligence, this defines a
context in which people will be surrounded by intelligent and intuitive inter-
faces embedded in everyday objects in an environment which will recognise and
respond to their presence in a way which is sensitive and context dependent and
which autonomously and intelligently adapts and responds to their needs [14]. As
well as houses, this consideration encompasses spaces such as “Smart Cities” [15–
17], strengthening concepts such as the Internet of Things (IoT) as is illustrated by
the development of the associated terminology shown in Fig. 12.2.
Fig. 12.2 Graphical representation of the evolution of terminology

12.2.1 New Services at Home
We are continually evolving our way of life, of work, of personal relationships and
so on. For example, it is now common to communicate (i.e. talk, share informa-
tion and so forth) almost every day with a wide community of individuals, some
of whom we may never have met. From a technological point of view, more and
more people seek an environment in which access to digital services from any-
where and at anytime is crucial. In this context, Information and Communications
Technologies (ICTs) need to be continuously upgraded to adapt to the individual
from two perspectives; services and technological infrastructure. For the former,
people demand new electronic services to service their need while the technologi-
cal infrastructure must support those services.
Traditionally, building automation and home automation have been associated
with safety, energy saving and comfort through the automation of services such
as lighting, environmental control and electrical appliances. In this context, it is
common to distinguish “personal security systems”, providing security services
directly to a user, from “central monitoring systems” connected to remote sites.
The progressive introduction of broadband to homes from the late 1990s then pro-
duced a change in the philosophy of service provision through the concept of the
Digital Home, eHome or iHome.
In recent years, the Ambient Intelligence paradigm has supported the develop-
ment of new smart devices integrated throughout networks to provide a service,
for instance to distribute and display multimedia. Moreover, the trend of adding
intelligence into devices is moving to wider environments as exemplified by the
Internet of Things and Smart Cities. The resulting Web of Everything will integrate
Smart Cyber-Physical Systems with the IoT to provide new forms of integrated
service [18, 19].
12.2.2 Designing the Smart Home
Consider the following scenario. A user in the study uses a PC on a Virtual

Personal Network, another user is in the living room, downloading and listening in
real time to music on the home stereo, yet another user is reading online newspa-
pers on a tablet and in the kitchen someone is doing the weekly shopping using the
electronic whiteboard on the refrigerator.
In this scenario, several electronic services are being used. However, the tech-
nology remains visible to the users and people have to learn to use the individ-
ual technologies in order to access the required services. The longer the learning
curve, the more visible the technology to the user and the more hidden the service.
Further, communication and control networks are often unconnected and use pro-
prietary systems.
Thus, in a current smart home the services and technology that provide support
can lack integration due to their spontaneous emergence in response to a perceived
need. As a consequence, services have tended to be specialised and the associated
technology is quite visible. However, the trend towards more intelligent systems
and components should enable environments to adapt to the user while hiding the
complexity of the underlying technology.
The resulting solutions have to consider two key perspectives, integration and
use. The former supports the provision of services irrespective of the technology
used and allows for the interconnection of a wide range of devices. The latter, the
user perspective, aims to support intelligent and transparent interaction by the user
with the technology.
12.2.2.1 Integration of Services and Technology
Systems integration is an important consideration for the future of home technolo-

gies. ICTs develop from a continually evolving technological base and it is impor-
tant that individual devices and systems are able to communicate with each other
to provide more complex services involving both co-operation and competition for
resources. In future, a distributed intelligence is likely to emerge from this interac-
tion to support both system configuration and operation.
In this context, it is possible to distinguish three different networks within a
smart home environment; namely control, multimedia and data, according to the
functions managed by the network:
• The control network provides the infrastructure for those services identified
with system automation and the management of simple commands and the
regulation of specific levels. Requirements include low cost, ease of installation
and reconfiguration, ease of expansion and fault tolerance. Devices connected to
this network are primarily sensors, actuators and controllers.
• The multimedia network provides support for the distribution of audio and video.
Requirements are related to the volume of data and the quality of service provi-
sion associated with the distributed audio and video data. Devices connected to
this network are televisions, HiFi equipment and other media-based items.
• Data networks were initially associated with the sharing of computing resources
such as files, programs and printers. With increased access to the Internet, the
data network must provide access to it from anywhere within the home environ-
ment. Requirements include high bandwidth and low cost. Connected devices
still include computers, printers, drives and scanners but also a wide range of
other data sharing devices such as tablets and smartphones as well as increasing
numbers of smart appliances.
The technology for each of these network types was initially proprietary or
designed specifically to provide the services associated with them. Today, borders
Fig. 12.3 Residential gateway
between multimedia and data networks are increasingly unclear as the bandwidth
evolution of the data network1 has resulted in the multimedia network becoming
effectively a subset of the data network.
From the perspective of the control network, the majority of current technol-
ogies remain those designed specifically to provide home automation. However,
convergence with data network technologies is increasing and control networks
now contain devices able to connect to both networks. Hence, the trend is for a
general convergence onto data network technologies.
Additionally, devices increasingly share information about themselves. For
instance, lamps can provide a range of information including their status, energy
consumption, hours of use and light levels, making them into a smart device.
Other appliances such as refrigerators can become a smart devices notifying us
(for example by email or SMS messaging) of shopping requirements, or even
placing the order online themselves. The trend is thus towards intelligent devices
which make use of network technologies to autonomously create larger, and more
complex, integrated systems.
The result is a complex scenario of integration structured around services
which are often specialised and strongly coupled to the technology that supports
them. Moreover, services usually operate in isolation unless they have been explic-
itly designed to cooperate. The gateway as a concept then plays an important role
in supporting communications between services and technologies through the con-
cept of the residential gateway of Fig. 12.3 to provide a single, flexible and intel-
ligent interface between external and internal networks.
Service and network levels were initially integrated with themselves and with
each other in a centralised structure with the emergence of the gateway. However, the
evolution of technology changes this centralised structure to a distributed structure
1Mainly technology structured around the IEEE 802 standard.

Fig. 12.4 A distributed system structure
in which the gateway supports any device with the necessary intelligence to enable it
to interact with others [20, 21]. In the same way, individual services should have the
ability to interact to provide a more elaborate service as suggested by Fig. 12.4.
There is therefore a technological challenge within ICT and the relationship
with the IoT to provide a standardised architecture to support the development of
new services. Horizon 2020 [22] commented that:
The biggest challenge will be to overcome the fragmentation of vertically oriented
closed systems, architectures and application areas and move towards open systems and
platforms that support multiple applications. The challenge for Europe is to capture the
benefits from developing consumer-oriented platforms that require a strong cooperation
between the telecom, hardware, software and service industries, to create and master inno-
vative Internet Ecosystems.
The above also implies the integration of smart components into cyber-physical
systems. Here, Horizon 2020 commented that:
Cyber-Physical Systems (CPS) refer to next generation embedded ICT systems that are
interconnected and collaborating including through the Internet of things, and providing
citizens and businesses with a wide range of innovative applications and services. These
are the ICT systems increasingly embedded in all types of artefacts making “smarter”,
more intelligent, more energy-efficient and more comfortable our transport systems, cars,
factories, hospitals, offices, homes, cities and personal devices.
Related to this new generation of components and systems, the challenge of

technological integration is to:
…. develop the next generations of smart systems technologies and solutions, based on
systemic miniaturisation and integration, of heterogeneous technologies ….
This is associated with the evolution of the Future Internet [23] focusing on
a redesign of the original client–server architecture to resolve issues of security,
trust and mobility. This future implementation has to meet:
the ever larger portfolio of business models, processes, applications/devices that have
to be supported, coupled with a rapidly growing number of application and societal
requirements.
12.2.2.2 Serving the Service
The user has to perceive the benefits of the service but not necessarily the technol-
ogy that support it. Although the integration of services and devices is an impor-
tant consideration, designing the user interface is potentially more challenging.
Interfaces have to be friendly and easy to use regardless of age and technology
skills. For example, a person has to interact with a home system to use the service,
define preferences, adapt services and so on. In this situation the concept of Maes
[24] that:
Autonomous agents are computational systems that inhabit some complex dynamic envi-
ronment, sense and act autonomously in this environment, and by doing so realize a set of
goals or tasks for which they are designed.
remains important for user interfaces.

The user interface has to emphasise autonomy while learning to perform tasks
for their users and providing proactive assistance as necessary. The user interface
could observe and monitor the actions performed by the user in order to learn,
suggest or perform an action. Additionally, the user interface should have the abil-
ity to adapt to the user.
Consider a person who wishes to set the house temperature to be 20 °C when
they arrive home from work. The most conventional solution is to programme the
time at which the heating system turns on, with obvious issues for early or late
return. A more evolved scenario would be the interaction of the user with the heat-
ing system via a website or mobile app. Taking this concept a stage further, the
heating system could independently communicate directly with (say) the user’s car
or relevant public transport system and autonomously turn on the heating based
on an estimated time of arrival, taking into account factors such as traffic, external
temperature and the (learned) heat transfer properties of the building.
The final solution requires a distributed intelligence within the system enabling
devices to cooperate or compete for resources and services. This implies smarter
devices incorporating predictive, reactive and cognitive capabilities.
Another important issue is the volume of information that home systems are
increasingly required to manage. Here, the predictive analysis of human behaviour
is an important issue to consider as is the predictive analysis of the home as entity
composed of different devices communicating with others both inside and external
to the home. According to the CONNECT forum [25]:
…. the integration of ‘Things’ as actors in the Internet via massive and innovative sensors,
actuators, and real-time reactivity will cause another order-of-magnitude data explosion
with challenges that we have yet to understand and deal with.
12.2.3 Home Systems Challenges
The most important goal is that of enabling individually intelligent devices to

effectively communicate and interact with the other devices and actors in the envi-
ronment. To provide intelligence at the device level, this could be seen as part of
the service layer of the device where the decisions are taken regarding the service
it provides. Alternatively, it could be seen as a layer which involves both services
and technology as it both provide services and uses the technology in different
ways depending on the requirements of the other devices in the environment.
At the technological level, advances are in two main directions, miniaturisation
and performance improvement. Within the IoT, reducing the size of the devices
to integrate them in essentially everything is an essential requirement, including
implantable devices as part of an eHealth and mHealth environment.
12.3 eHealth and mHealth
In the introduction to the report ‘Good health adds life to years: Global brief for
World Health Day 2012’ Dr. Margaret Chan, the then Director-General of the
World Health Organization,2 wrote that [26]:
Population ageing is a global phenomenon that is both inevitable and predictable. It will
change society at many levels and in complex ways, creating both challenges and oppor-
tunities ….. This great demographic challenge of the first half of the 21st century therefore
demands a public health response….
In another context, the World Economic Forum in their series of Global Risks
Reports [27] has consistently identified mismanagement of population ageing as
a high likelihood, high impact area lying on the societal axis of their analysis.
This challenge of an ageing society is not only a global issue, as illustrated by
Fig. 12.5a, b, but also one that is accelerating [28].
Globally, the effective delivery of all aspects of healthcare is an increasing pri-
ority, and the World Health Organisation commented that [29]:
…. as long as the acute care model dominates health care systems, health care expendi-
tures will continue to escalate, but improvements in populations’ health status will not.
2Appointed 2007 and re-appointed for a further 5-year term in 2012.

Fig. 12.5 The challenge of ageing. a Growth in population aged 60+ years. b Dependency
ratios—defined as the number of people aged over 65 for every 100 people in the age range 15–64
Further, the EU Commission in 2009 noted that [30]:

In order to limit the expected increase in public expenditure, policy measures which can
either reduce disability, limit the need for formal care amongst elderly citizens with dis-
abilities, favour formal care provision at home rather than in institutions or, more gener-
ally, improve the cost-effectiveness of long-term care provision, e.g. through introduction
of eHealth and telecare must be developed.
These factors, together with the associated demands on physical, human and
social resources, has led to a consideration of wide-ranging eHealth strategies
deploying advanced ICTs as a means of providing the desired and required levels
of support. However, the evidence base remains relatively weak and Demiris and
Hensel in their 2008 systematic review of health related smart home applications
[31] stated that:
…. in spite of the growing number of initiatives in this area, the field is in relatively early
stages and is currently lacking an extensive body of evidence.
and there has been little since to suggest a significant change in this position.
In practice, evaluation has to date tended to be based on relatively limited
evidence, often structured around the extrapolation of relatively small data sets,
which themselves are often concentrated around a focussed application or a
selected group of participants. It is therefore suggested that such evaluation as
does exist can perhaps best be categorised as trials aimed at establishing the per-
formance of specific system components rather than establishing their functional
and operational integration at the system level, for whom and in what circum-
stances [32, 33]. The effect has thus been that to date installations have essentially
been experiments, and need to be considered and evaluated as such.
In the absence of a wider integration of such data as is available, access to
which may well have commercial implications, this position of sparse data and
lack of confirmation is likely to be the case for the immediate future.
There is also the concern that within the overall context of eHealth there has
been an inevitable, and to a degree understandable, compartmentalisation of tech-
nologies and applications in order to integrate them within conventional healthcare
structures and organisations. Thus for instance, physiological monitoring is often
seen as a constituent component of telehealth, but not of telecare, whereas from
both a technical perspective, and perhaps more importantly a user perspective,
they form part of a continuum of applicable technologies.
It is also the case that there has been a significant shift in the nature of technol-
ogy since telecare, telehealth and telemedicine systems began development in the
late 1980s and early 1990s. Of these, perhaps the most significant has been the
evolution of smart objects and their interconnection through the medium of the
IoT. This level of connectivity is illustrated by Fig. 12.6 which shows the rise in
the number of connected devices per person.
The result is a combination of technology push in the development of new and
novel forms of sensing, cloud computing, near field communications, smart com-
munications, adaptive and emotive computing, machine ethics, and user pull driv-
ing demands for increased service provision.
Developments such as the IoT and Cyber-Physical Systems imply the large-
scale interconnection of a range of smart, and essentially mechatronic, objects to
service information [34, 35]. This, however, represents a paradigm shift in sys-
tems development from an environment in which information is used to service
artefacts to one in which (smart) artefacts are used to service information. This in
Fig. 12.6 Growth in connectivity
turn implies a change in the way in which systems are viewed as being relatively
inflexible and task oriented to highly flexible goal-oriented entities whose role can
easily be changed by reconfiguring the connected smart objects. In such an envi-
ronment, connectivity, and the ability to configure user interfaces to suit individual
requirement and need becomes paramount to system functionality.
Consider an eHealth system structured around the following core functions:
• The monitoring and analysis of activity to detect and respond to anomalies or
indicators of changed status, and hence a change in need.
• Physiological sensing and the recording of symptoms.
• The monitoring of emotion and its integration with other forms of behavioural
data.
• The recording of specific observational data related to the monitored individual.
• The use of smart interfaces to provide the link between the individual and the
system.
• The use of the Cloud as a data storage and transfer medium.
In the context of the IoT this functionality can then be associated with appropriate
groups or clusters of smart objects to provide the connectivity.
Table 12.1 Characteristics of technological and other forms of lock-in

Characteristic
• Consumption of resources continues to increase even after changes which would seem to
permit reductions
• The take-up of a new technology is a function of the ‘inertia’, as expressed for instance
through market position, of the existing technologies and systems
• Technologies become embedded within political, social and economic systems such as mar-
kets, patterns of consumer demand, systems of regulation and infrastructure
• Institutional lock-in arising from existing interests and an adherence to convention
• Inferior designs become fixed in use by a process in which circumstance are often as important
as the design itself
• Established technologies often show economies of scale, and replacement with new designs
and processes can incur significant structural and resource costs at the introductory stage, even
where the long term consumption of resources, functionality, environmental impact and costs
are likely to be superior
12.3.1 System Design Issues in eHealth and mHealth
A substantial background of research has defined the current state of knowledge

and provided an understanding of the issues involved. However, there has perhaps
been a trend towards increasingly complex technological solutions [36], potentially
resulting in a lack of focus on the user. Approaches which compartmentalise sys-
tem elements are inevitably vulnerable to a form of ‘lock-in’, in which self-rein-
forcing barriers act to inhibit change or prevent the uptake of new technologies or
the integration of such technologies within established systems [37]. The character-
istics of such a lock-in are summarised in Table 12.1, and are often associated with
the levels of research, time and investment that have gone into a system or technol-
ogy, and the resulting, and entirely understandable, desire to make this work.
To take advantage of developments in technology, it is therefore argued that
there needs to be a more open approach to system design and implementation
along the lines of the Open Innovation strategy as outlined by Chesbrough [38].
12.3.1.1 Sensors
The increasing availability of new generations of initially wearable, but almost

certainly ultimately implantable, sensors capable of providing information on
a range of physiological signs such as pulse, respiration, body temperature and
movement needs to be accommodated within the next generation of eHealth sys-
tems. While the data from these, as for instance relating to movement, could be
correlated with activity data derived from PIRs, they could also act independently
in establishing, on a clinical basis, the requirements for an emergency response.
Specifically, the introduction of wearable, and eventually implantable, sensors

supports a separation of the functions of monitoring and emergency response in a
way which is linked directly to the individual. It is therefore argued that this form
of sensing has the potential to play a significant role within eHealth and mHealth
in a number of different roles, including [39]:
• The direct monitoring of a range of physiological parameters.
• The separation of functions such as general behavioural monitoring and emer-
gency response with physiological data being used to supplement data from the
behaviour monitoring system.
• The ability to extend the support for an individual from the home environment
to the wider environment through the linking of the body hub or implanted sen-
sors to mobile communications networks which would provide a continuous
monitoring of the appropriate data.
12.3.1.2 mHealth
The coming together of the technologies of worn or implantable sensors with

mobile communications affords an opportunity to expand the coverage to the envi-
ronment as a whole, enhancing the ability of an individual to move between the
home and the wider environment. Thus, an individual with implanted physiologi-
cal sensors has an application loaded onto a smart phone to enable it to serve as
their personal information node. This then allows their data to be monitored in
both the home and wider environments.
In normal use, this data will be integrated with home derived data such as the
recording of the use of space as part of the general monitoring process. Should an
abnormal condition arise then depending on the nature of the abnormality, a num-
ber of actions could result, including:
• Messages passed to the care team for appropriate follow-up.
• The user advised of any immediate actions to take.
• An emergency response initiated if appropriate.
Should this condition be such as to cause, say, the individual concerned to be
unconscious when the emergency response team arrive, the fact that they are
linked to their health record would enable the response team, with proper safe-
guards in place, to access information such as known medical conditions and
medication to enable a more effective response than might otherwise be the
case.
Mobile health or mHealth-related applications structured around the use of
mobile communications as a specific element within eHealth thus offer great
promise. Applications include the collection of data, dissemination of information,
and patients, real-time monitoring and related issues [40, 41].
12.3.1.3 Standards
Standards and protocols are competing for the home networking market are
largely based around the IEEE 802.11 standard and its range of amendments.
Their use for telecare applications is then dependent on systems providers adopt-
ing appropriate standards to allow for the networking of a range of devices.
However, and for understandable commercial reasons, systems suppliers are at
times reluctant to adopt an open systems approach, which would allow devices
from a range of suppliers and providers to be integrated onto a single home net-
work and which put in place appropriate safeguards for the handling of the health-
related data generated [42].
In relation to eHealth systems, the ISO/IEEE 11073 Personal Health Data
(PHD) standards [43] aim to support:
• The provision of real-time plug-and-play interoperability for medical, health-
care and wellness devices.
• The facilitation of the efficient exchange of care device data, acquired at the
point-of-care, in all care environments.
12.3.1.4 Informatics, Data Security, Ethics and a Knowledge Economy
The integration and management of healthcare-related data from all sources pre-
sents a major informatics challenge in ensuring the robust and secure control of
an individual’s data whilst allowing appropriate access to that data [44]. Indeed,
resolving questions of data and information security is likely to be a major issue in
developing personal health databases to achieve the necessary levels of user confi-
dence while ensuring appropriate access.
Other issues of growing importance are the ability to use an integrated health
informatics system, sometimes referred to as a ‘Learning Health System’ [45],
in which the ability to use the information residing in an integrated health infor-
matics system, and particular patient data, can be used in support of the planning,
organisation and management of a health system, for instance through the identifi-
cation of trends and patterns to enable earlier interventions to take place.
Such systems envisage an infrastructure along the lines of that shown in
Fig. 12.7 in which related information can move freely between the strategic
groupings associated with the provision, development and management of health-
care while relating back to the individual. Again, however, there are significant
issues of security and confidentiality. In particular, the ensuring of patient ano-
nymity while enabling the relevant data to be accessed for strategic purposes.
Studies of the potential of cloud computing in healthcare [46, 47] have served to
highlight and identify problems associated with the maintenance of data security
within a shared environment whilst also suggesting pointers to potential solutions.
Fig. 12.7 eHealth/mHealth infrastructure
Fig. 12.8 Knowledge
economy
Further, discussions regarding the general access to, and use of, anonymous
data have confirmed the significant levels of concern associated with providing
access to such data, even when it is recognised that the application is benign.
All of the above have ethical, both human- and machine-related, implications
which will need to be addressed if the approach is to become adopted. In particu-
lar, there are the concerns of allocating the responsibility for the well-being of an
individual to an autonomous computer based system which makes decisions on
their behalf as to their state of health.
Further, such an infrastructure must be positioned within the context of an
evolving knowledge economy as suggested by Fig. 12.8 within which the data and
information relating to individuals may well be seen as having significant con-

textual, or indeed financial, value. Thus, concerns have been expressed that such
data may be traded and used to prohibit access to certain levels of provision. The
ethical issues associated with such information transactions, and their balancing
to maximise the return to the individual while optimising the management of their
healthcare provision will need to be resolved if eHealth and mHealth are to impact
on such provision [48, 49].
12.3.2 eHealth and mHealth Challenges
In an environment seeing a rapid growth in the number of older people places sig-
nificant additional demands on resources, including infrastructure resources such
as housing, communications and transport. Further, many of these older individu-
als wish to live as independently as possible for as long as possible, posing soci-
etal challenges in ensuring access and mobility whilst preventing trends such as
increasing urbanisation and the depopulation of rural areas, as for instance the
Highlands and Islands’ in the UK.
The underlying vision must therefore be one of an environment in which stake-
holder needs are met through a sustainable organisation and the structuring of both
the physical environment and the information environment to meet the changing
needs of an ageing population.
Thus for instance, mobility must be considered not just as an ability to move
within the physical environment, with all that that implies, but also mobility within
the information environment, for instance through the deployment of new and novel
approaches to interfaces and visualisation. It is argued that such enhanced mobility
within the information environment then acts to support physical mobility, for instance
through developments in mobile healthcare (mHealth) to allow aspects and elements
of telecare to move with the individual rather than be fixed to their home environment.
This overarching vision of a healthcare infrastructure which integrates the
physical and the information environments to support the user in turn implies the
need for sustainable solutions which maximise benefits whilst optimising the use
of resources in each of the short, medium and long terms. These solutions must
address and support issues such as
• The balancing of the level of provision between urban and rural communities
whilst recognising the needs of such communities.
• The nature of the housing stock and the balance between new build and refit or
refurbishment.
• The ability to effectively assess need, specifically within the home environment
rather than a laboratory.
• Means of capturing new and novel forms of data such as observational data.
• The design strategies to be adopted in relation to each and all of these issues.
• The nature of the tools to be used to support the effective assessment of the
impact of change.
It must also be recognised that many current systems have over time been the sub-
ject of evaluation, review and indeed change. This has in turn often resulted in an
interest, and indeed in some cases an investment in maintaining the status quo,
resulting in a potential for a degree of technological lock-in which acts to inhibit
the introduction of new concepts, methods and ideas. Meeting these and related
challenges is fundamental to managing an ageing population.
Key issues for debate are therefore suggested as being:
• Establishment of user needs and requirements through interaction with the full
range of stakeholder groups. This is an aspect of the process that is too often
neglected and requires the development and implementation of conflict resolu-
tion strategies to ensure a balanced outcome.
• Identification of resources and of the interactions, both levels and forms, between
different types of resource; i.e., human, financial, societal, infrastructure, etc.
• Infrastructure issues impacting upon both perceived and real issues of access
and mobility.
• Sustainability issues impacting upon access to and availability of resources and
including economic sustainability.
• Issues of user interaction and the incorporation of the necessary communica-
tions infrastructures within the built environment.
• Definition and evaluation of methods and techniques to carry out user evalua-
tions in both the laboratory and the home environment.
• Design strategies and methods and the exporting of these to the relevant stake-
holder groups.
• The development and implementation of decision support and related tools to
inform decision-makers as to options and outcomes.
• The role of the IoT as a means of integrating a range of smart and mechatronic
objects within that infrastructure.
The question must also be asked as to why, despite advances in technology that
have taken place over the past 20 years, have there been no validated large-scale
eHealth installations taking advantages of those technologies. It is suggested here
that this is because the development of the systems required to make effective
use of the technologies have lagged behind the development of the technologies.
Requirements here are for:
• The deployment of the techniques of data mining and knowledge extraction
as applied to large data sets and the dissemination of that knowledge to the
individual.
• The restructuring of functional, organisational and operational systems and pro-
cedure within the overall context of healthcare provision to make effective use
of new knowledge being generated.
• The future proofing, as far as is practical or possible, of such systems to take
account of future developments.
• Ethical and security issues associated with the management of the growing vol-
ume of both generic and user-specific data.
12.4 Conclusions
A major focus of attention in the development of future Smart Homes is the ability
to make the underlying technology as non-invasive as possible. Thus in relation to
the user interface, computer vision and speech recognition offer many possibili-
ties, but require improvements over current solutions. Included here is the need to
achieve new and novel solutions that encompass machine intelligence to extract
useful data from visual and audible information together with the communication
protocols necessary to secure and fast communications. Developments up to the
time of writing in these areas include Apple Siri,3 Microsoft Cortana4 for speech
recognition and Microsoft Kinect,5 Leap Motion6 and Sentry Eye Tracker7 for
motion recognition based on computer vision.
While computer vision has the potential to simplify and improve the interaction
of the users with a smart environment there still exist a number of challenges to be
resolved, including:
• Facial expressions in adverse lighting conditions, for instance during the night
when the light levels are low.
• The environment may host several people and the system may be required to
extract both individual and global information, for instance to lead people to the
safest exit in the event of fire.
• Provision of methods and means which ensure privacy, particularly where the
system may be required to respond differently to the differing needs of different
users.
Wearable devices also have the potential to support location independent services.
These could include:
• Detecting and responding to dangerous situations such as a hole in the street or
approaching traffic.
• Route guidance in unknown locations.
• Support in emergencies, for instance by informing users as to potential actions
while autonomously communicating with emergency services.
In a wider system context, challenges include:
• The ability to analyse and respond to the level of activity in a street to autono-
mously regulate the lighting, traffic signals and other elements of the environ-
ment to optimise factors such as safety and energy consumption.
3www.apple.com/ios/siri (accessed 8 October 2015).

4www.windowsphone.com/en-US/how-to/wp8/cortana/meet-cortana (accessed 8 October 2015).
5http://dev.windows.com/en-us/kinect (accessed 8 October 2015).
6www.leapmotion.com (accessed 8 October 2015).
7http://steelseries.com/gaming-controllers/sentry-gaming-eye-tracker (accessed 8 October 2015).
• Detect and report on damage to infrastructure to support pre-emptive

maintenance.
• Detecting and responding to dangerous or hazardous situations and acting to
mitigate these.
• The provision of design tools to support the design of integrated systems based
around the presence of numbers of smart devices.
• Ensuring user privacy.
The chapter has set out to try, through the medium of the Smart House and
eHealth technologies in particular, to suggest how the increasing availability of
smart devices and systems will impact on the way in which individuals react to
and interact with their environment, and to identify some key issues for which res-
olution is required in order that the user potential may be achieved.
References
1. Alam MR, Reaz MBI, Ali MAM (2012) A review of smart homes—past, present, and future.
IEEE Trans Syst Man Cybern Part C Appl Rev 42(6):1190–1203
2. Launey RO et al (1992) Expandable home automation system. US Patent No 5,086,385
3. Humphries LS et al (1997) Home automation system. US Patent No 5,621,662
4. Mun YS (1996) Home automation system having user controlled definition function. US
Patent No 5,579,221
5. Henry SG, Tsunemura K, Kiyoshi IHA (1985) Garage door operator and method of assem-
bling. US Patent 4,538,661
6. Bly S, Schilit B, McDonald DW, Rosario B, Saint-Hilaire Y (2006) Broken expectations in
the digital home. In: CHI’06 extended abstracts on human factors in computing systems.
ACM, pp 568–573
7. Kirchhof M, Linz S (2005) Component-based development of Web-enabled eHome services.
Pers Ubiquit Comput 9(5):323–332
8. Norbisrath U, Armac I, Retkowitz D, Salumaa P (2006) Modeling eHome systems. In:
Proceedings of 4th international workshop middleware for pervasive and ad-hoc computing
(MPAC2006), p 4
9. Dobrev P, Famolari D, Kurzke C, Miller B (2002) Device and service discovery in home net-
works with OSGi. IEEE Commun Mag 40(8):86–92
10. Rose B (2001) Home networks: a standards perspective. IEEE Commun Mag 39(12):78–85
11. Satyanarayanan M (2001) Pervasive computing: vision and challenges. IEEE Pers Commun
8(4):10–17
12. Schmidt A (2005) Interactive context-aware systems interacting with ambient intelligence.
In: Ambient intelligence, p 159
13. Bohn J, Coroamă V, Langheinrich M, Mattern F, Rohs M (2005) Social, economic, and ethi-
cal implications of ambient intelligence and ubiquitous computing. In: Ambient intelligence,
pp 5–29
14. Aarts E, Collier R, van Loenen E, de Ruyter B (2003) Ambient intelligence. EUSAI 2003,
vol 2875, p 432
15. Hollands RG (2008) Will the real smart city please stand up? Intelligent, progressive or
entrepreneurial? City 12(3):303–320
16. Su K, Li J, Fu H (2011) Smart city and the applications. In: IEEE international conference on
electronics, communications and control (ICECC 2011), pp 1028–1031
17. Morandi C, Rolando A, Di Vita S (2015) From smart city to smart region, digital services for
an Internet of Places. Springer, Berlin
18. Thomas I, Gaide S, Hug M (2013), Composite business ecosystems for the web of every-
thing: using cloud platforms for web convergence. In: 7th IEEE international conference on
innovative mobile and internet services in ubiquitous computing (IMIS), pp 535–540
19. Pendyala VS, Shim SSY, Bussler C (2015) The web that extends beyond the world.
Computer 48(5):18–25
54(15):2787–2805
21. Perera C, Zaslavsky A, Christen P, Georgakopoulos D (2014) Context aware computing for
the Internet of Things: a survey. IEEE Commun Surv Tutorials 16(1):414–454
22. European Commission, HORIZON 2020, Work Programme 2014, Leadership in enabling
and industrial technologies (Information and Communication Technologies) @ http://ec.
europa.eu/research/participants/data/ref/h2020/wp/2014_2015/main/h2020-wp1415-leit-ict_
en.pdf. Accessed 8 Oct 2015
23. Domingue J, Galis A, Gavras A, Zahariadis T, Lambert D, Cleary F, Schaffers H (eds) (2011)
The future internet, future internet assembly 2011: achievements and technological promises.
Springer, Berlin
24. Maes P (1995) Artificial life meets entertainment: life like autonomous agents. Commun
ACM 38(11):108–114
25. CONNECT Advisory Forum (2014) @ http://ec.europa.eu/digital-agenda/en/connect-advi-
sory-forum-working-groups. Accessed 18 Oct 2015
26. World Health Organisation (2012) Good health adds life to years: global brief for World
Health Day 2012 @ www.who.int/ageing/publications/whd2012_global_brief/en/index.html.
Accessed 26 May 2015
27. www.weforum.org. Accessed 26 May 2015
28. Kinsella K, He W (2009) An aging world: 2008. National Institute on Aging and US Census
Bureau @ www.census.gov/prod/2009pubs/p95-09-1.pdf. Accessed 26 May 2015
29. World Health Organisation (2002) Innovative care for chronic conditions: building blocks for
action: global report @ www.who.int/chp/knowledge/publications/icccreport/en/. Accessed
26 May 2015
30. Commission of the European Communities (2009) Dealing with the impact of an ageing pop-
ulation in the EU, COM (2009), 180/4 @ http://ec.europa.eu/economy_finance/publications/
publication14992_en.pdf. Accessed 26 May 2015
31. Demiris G, Hensel BK (2008) Technologies for an Ageing Society: a systematic review of
‘Smart Home’ applications. IMIA Yearb Med Inf 47(Suppl 1):33–40
32. Rahimpoura M, Lovell NH, Celler BG, McCormick J (2008) Patients’ perceptions of a home
telecare system. Int J Med Inf 77:486–498
33. Boger J, Mihailidis A (2011) The future of intelligent assistive technologies for cogni-
tion: devices under development to support independent living and aging with choice.
NeuroRehabilitation 28:271–280
54:2787–2805
35. Shi J, Wan J, Yan H, Suo H (2011) A survey of cyber-physical systems. In: Proceedings of
international conference on wireless communications and signal processing, pp 1–6
36. Brownsell S, Bradley D, Blackburn S, Cardineux F, Hawley M (2011) A systematic review of
lifestyle monitoring technologies. J Telemedicine Telecare 17:185–189
37. Foxon TJ (2007) Technological lock-in and the role of innovation. In: Atkinson G, Dietz S,
Neumayer E (eds) Handbook of sustainable development. EE Publishing, pp 140–151
38. Chesbrough HW (2003) Open innovation: the new imperative for creating and profiting from
technology. Harvard Business Press, Boston
39. Pantelopoulos A, Bourbakis NG (2010) A survey on wearable sensor-based systems for
health monitoring and prognosis. IEEE Trans Syst Man Cybern Part C Appl Rev 40(1):1–12
40. Sama PR, Eapen ZJ, Weinfurt KP, Shah BR, Schulman KA (2014) An evaluation of mobile
health application tools. JMIR mHealth uHealth 2(2):e19 @ http://mhealth.jmir.org/2014/2/
e19/. Accessed 28 May 2015
41. Vital Wave Consulting (2009) mHealth for development: the opportunity of mobile technol-
ogy for healthcare in the developing world (PDF). United Nations Foundation & Vodafone
Foundation @ www.vitalwaveconsulting.com/pdf/2011/mHealth.pdf. Accessed 28 May 2015
42. 66th World Health Assembly (2013) eHealth standardization and interoperability, WHA66.24
(Agenda item 17.5) @ http://apps.who.int/gb/ebwha/pdf_files/WHA66/A66_R24-en.pdf.
Accessed 5 Nov 2015
43. Song J, Kong XY, Wu HC, Wei Y (2013) Integrate personal healthcare devices into smart
home with ISO/IEEE 11073 PHD standard. Appl Mech Mater 394:487–491
44. Hovenga EJS, Kidd MR, Garde S, Hullin Lucay Cossio C (eds) (2010) Health informatics:
an overview. IOS Press Amsterdam, 520 pp
45. Friedman CP, Wong AK, Blumenthal D (2010) Achieving a nationwide learning health sys-
tem. Sci Transl Med 2(57):1–3
46. Rolim CO, Koch FL, Westphall CB, Werner J, Fracalossi A, Salvador GS (2010) A cloud
computing solution for patient’s data collection in Health Care Institutions. In: 2nd interna-
tional conference on eHealth, telemedicine and social medicine, pp 95–99
47. Guo L, Chen F, Chen L, Tang X (2010) The building of cloud computing environment for
E-Health. In: international conference on E-Health networking, digital ecosystems and tech-
nologies, pp 89–92
48. George C, Whitehouse D, Duquenoy P (eds) (2012) eHealth: legal, ethical and governance
challenges. Springer, Berlin
49. Anderson M, Anderson SL (eds) (2011) Machine ethics. Cambridge University Press,
Cambridge
Chapter 13
The Changing Landscape of Enterprise
ICT—Responding to New Possibilities
and User Demands
Christopher Milne and Steve Watt
13.1 An ICT Practitioners Perspective
As Information and Communications Technology (ICT) practitioners and manag-

ers in a higher education institution, our work essentially requires the integration
of a diverse range of elements to develop, implement, maintain and run a range
of systems and services. These elements include people (internal and external);
responses to legislation and regulation; organisational and user requirements; cus-
tomers (internal and external); hardware; software; data; internal infrastructure
and external (regional and national) infrastructure (e.g. the SuperJANET network).
These elements are integrated at both systems and organisational levels, to pro-
vide what can be described as commoditised ICT systems and services. However,
focus is steadily shifting from delivering commodity ICT systems and services
to driving and sustaining organisational growth through business transforma-
tion by business process reengineering and structured around a range of enabling
technologies, many of which incorporate their own integral software and sys-
tems. We therefore provide a different perspective on the opportunities and chal-
lenges that are likely to form part of the mechatronics futures, by reflecting on the
key features of the changing environment that we operate in and the challenges
of integrating the elements noted above to deliver services and secure business
transformation.
C. Milne · S. Watt (*)

University of St Andrews, St Andrews, UK
C. Milne

DOI 10.1007/978-3-319-32156-1_13
202 C. Milne and S. Watt
13.1.1 Developing and Delivering Services In-House

Is Suboptimal
The environment and methods through which enterprise ICT systems are estab-
lished to deliver services to end-users have changed and are continuing to change.
Traditionally, the ICT function was predominately an in-house operation in terms
of resourcing, approach and scope. Organisations invested to recruit, develop and
retain a skills base necessary to deliver services provided by the in-house ICT
function. The in-house work force would thus consist of core workers [1], that is
permanently contracted employees, who (other than planned or unplanned leave)
were readily available to the organisation, supplemented by non-core workers such
as contractors, when specialist or unique skills were required on an ad hoc basis.
Organisations sought ownership of their core in-house ICT infrastructure, with
some equipment leasing taking place on the periphery. In-house services, as the
phrase suggests, more often than not, were delivered within the geographical and
physical boundaries of the organisation.
By their very nature, the majority of in-house functions are inward look-
ing in so far as they are normally created to serve organisational requirements.
Connectivity between in-house functions is the norm, with external networking
being less prevalent and/or structured [2]. In a world where options for connec-
tivity and collaboration in all forms are increasing exponentially; where resources
are more scarce, or expensive and systems and services require a higher degree
of integration—designing and providing systems and services with in-house work-
ing as the default philosophy is open to significant challenge. For instance, con-
sider the collaboration across organisations to create the iPod and the associated
changes in licensing necessary to allow end-users to access and download copy-
right material in different ways [3].
From an economic perspective, in many instances, designing and providing ser-
vices based on in-house philosophies is now becoming a suboptimal approach to
systems delivery and operation.
13.1.2 Cloud Computing
Advances in cloud computing are changing the enterprise ICT landscape, and in
particular organisations are no longer bound by in-house provision. This is an
exciting and welcome development, resulting in significant momentum in cloud
adoption as a way of delivering services that once would be delivered by a combi-
nation of in-house development, hosting and support.
Expending resource to deliver services by establishing, implementing, main-
taining, securing, developing and retaining, an ICT enterprise infrastructure that is
part of an enterprise (in both physical and balance sheet perspectives) in a number
of areas, no longer makes any business sense. Indeed, following this traditional
13 The Changing Landscape of Enterprise ICT … 203
service model can limit the capacity of an enterprise’s ICT function to create value
and competitive advantage. Cloud providers have developed the infrastructure
economies of scale and new transformational business models. They can absorb
new customers rapidly, providing services as modules, a form of enterprise plug
and play.
In the higher education sector, student email services are a relevant example.
Here, a significant number of UK universities now contract with cloud provid-
ers such as Microsoft or Google for those services [4], freeing in-house resource
which can be used to deliver other services, or be returned to the business, adds
greater value.
Cloud-based models of ICT service delivery require a new, and notably diverse,
skills set—partner identification, contract and business relationship management,
information governance, data architecture and developing back-out plans. How do
you secure enterprise data from a cloud provider, returning this to the enterprise
and/or a different infrastructure partner in the future?
This is changing the boundaries of systems analysis and design at the enterprise
level, the Chief Technology Officer (CTO) or Chief Information Officer (CIO)
focus is no longer on devices and device management. Infrastructure is the pre-
serve of the cloud provider. Enterprise systems analysis and design now involves
risk analysis, contract and data protection law and establishing a firm partnership.
The risk analysis process includes due diligence around the cloud provider and an
evaluation of all of the risks associated with both the intended current and future
use cases and provider. This can include data sovereignty, location, security and
exit options.
There is more value and scope to create competitive advantage by identifying
a suitable partner and accessing third-party infrastructure available in the cloud.
This creates scope to free in-house resource to create value for the enterprise else-
where such as outsourced electronic mail, which is a relatively common example
of Software as a Service (SaaS) and High Performance Computing becoming a
more commonplace example of Infrastructure as a Service (IaaS). One can appre-
ciate the value of this more by considering the outsourcing of mail scanning and
filtering where a third-party supplier has access to a far larger database of threats
than an on premise service, thus providing a better solution.
13.1.3 Focus Is Shifting from Device Management

to Data Management
Our dependency on the Internet and on the many devices that connect to it con-
tinues to increase year on year. An “always on” society demands greater con-
nectivity in a ubiquitous manner. Fuelling this further is the increase in common
devices, such as digital cameras, televisions which now have Internet connectivity.
Traditionally, such devices were used in isolation. Their purpose and functionality
is now extending dramatically and a broader perspective on networking and data

transmission is creating new opportunities and challenges along with increased
demands from users and customers.
In less than a generation, the term “online” has moved from the lexicon of elec-
tronic engineering and computer science related disciplines to become embed-
ded in almost every facet of daily life. Fewer activities now take place exclusively
offline, or without an element of online interaction. People continually demand
more from their online experiences—it is not good enough that a device works
reliably and that connection to the Internet (or other networks) is seamless and
near instantaneous. People are seeking higher levels of integration between
devices and all pertinent data sources as for instance through citizen portals, stu-
dent portals and online one-stop shops. Such services also need to be platform
agnostic and responsive to cater for the range and variety of access methods and
devices.
While devices and services are important, the value and the role of data should
not be overlooked. Many organisations that were technology or device centric are
changing their business models—selling data and/or data integration services is in
some instances becoming more important than trading hardware.
In October 2011, Hewlett-Packard (HP) acquired Autonomy Corporation
plc—a specialist in the analysis of unstructured “big data.” While the financial
prudence of the £7.4 billion takeover has been the subject of much comment [5],
HP’s purchase of Autonomy signalled a significant strategic shift towards revenue
creation from processing of structured and unstructured data. That shift has contin-
ued. In November 2015, HP split into two separate companies: HP Enterprise will
focus on supplying corporations with software, services, and hardware with HP
Inc providing computers and printers. In 2014, the Microsoft CEO, Satya Nadella,
in an email to all employees outlined the future direction of the company, signal-
ling a shift in focus from devices and services towards mobile and cloud [data] [6].
13.1.4 Security
One of the greatest challenges which enterprises have recently come to face is
that of cyber security. In the UK cyber security is considered as a Tier 1 threat,
the same threat level as terrorism or natural disaster. With Gartner predicting that
over 75 % of business will be “digital” by 2020 one can quickly identify the likely
impact on both the enterprise and the end-user. It is therefore hardly a coincidence
that the annual World Economic Forum Risk Report [7] has consistently identified
cyber security and associated factors as a major and high impact risk area.
Large integrated systems like those which are found in most enterprises are
complex systems of people, processes and technology [8]. They can also be con-
sidered as adaptive even as far as to say they can exhibit characteristics of having
a life of their own. This has direct links back to the study of biological systems
in that often their behaviour under certain circumstances is unpredictable. Should
a security breach occur, it is very difficult to understand the implications on the

many smaller systems which then integrate to form the enterprise system.
Enterprise local area networks (LANs) must now support a range of complex
systems ranging from traditional network connected devices to embedded sys-
tems within intruder alarm systems, access control, building management sys-
tems, energy production, traffic control systems and many other instances. Each
of these devices presents its own challenges especially where IP connectivity has
been added by the manufacturer beyond the initial design phase. Recent work by
Hyungjun [9] has highlighted key steps which must be taken to provide sound
security in an IP-based SCADA environment. A well-documented example of
such issues is in the area of SCADA where a number of security breaches have
occurred as a result of poorly written or maintained software on such devices [10].
The nature of such systems often requires a non-traditional approach to secu-
rity to be taken. It is also the case that the traditional approach to securing our
networks is no longer acceptable—an issue that will only further be exacerbated
as we move forward. Securing networks now requires focussing effort more on
securing what is important to us rather than trying to implement a lockdown
approach and trying to protect everything. Historically, a firewall between the
LAN and the Internet provided an organisation with a level of protection from
external threats by (theoretically) allowing only legitimate network traffic to pass
through. This approach is no longer workable as so many devices located within
the LAN require access to Internet resources. A focus on data security is now
needed which recognises that the boundaries between the LAN and the Wide Area
Network (WAN) are blurred. This is often referred to as the “disappearing perim-
eter” [11]. Every single endpoint now needs to be considered as a potential stag-
ing area for attack—with endpoints being a mixture of corporately owned assets,
worker assets or visitor assets in the context of an enterprise. There will be an end-
point security resurgence as a result of the Internet of Things (IoT).
Focus should therefore be around securing the data rather than the organisation
as it is data that is the ultimate target. As data can reside on physical servers, vir-
tual servers or cloud-based services, the solutions required are different but must
be managed from within the same management interface if a holistic approach is
to be taken.
The impact of security on usability is another area which has also been identi-
fied as a barrier to adoption. A recent example in the education environment was
the implementation of endpoint encryption. This was considered necessary to
protect personal data and a policy agreed to implement a solution which would
encrypt the data stored on portable ICT devices such as smart phones, laptops, tab-
lets etc. The adoption of this solution was slow, and is still to this date seen as
an impediment by users. Similar examples can be found where usability is often
given more consideration than security, and indeed security is often omitted to
ensure that the end-user experience is maintained—often resulting in disaster [11].
One of the many challenges within an open environment is identifying where
legitimate user credentials are being used to carry out malicious acts.
13.1.5 Strategic Partnership Working
According to Prahalad and Krishnan [12], innovation is derived more successfully

using a group of industry partners and customers rather than in the traditional way
where one organisation would develop the full ecosystem themselves and involve
the customer thereafter as a consumer of the end product and/or service. Bringing
together the collective core competencies from a number of partners provides a
better platform for innovation. There are some parallels to this to what we have
outlined earlier in identifying that a future delivery model for IT Services will
be better optimised if it is done through strategic partnering in a very open way.
Prahalad and Krishnan refer to this as “co-created value”. Parallels also exist with
the work of Chesborough [13] around the concept of open innovation.
This approach is becoming more common and will inform organisational
design as a new set of roles will be required to manage such relationships. This
will include Relationship Managers and Contract Managers who specialise in
contractual and supplier management rather than traditional management of infra-
structure and in-house IT departments. This shift is also being seen when tradi-
tional enterprises move towards the use of commodity cloud services—whether
public, private or hybrid resulting in a new skill set being required with strong
engineering and integration skills.
Looking forward we will see a steady shift from CIO roles to CEO roles as
Information Technology becomes a more integral part of corporate strategy. In
summary, such roles will be focussed on solutions rather than the intricacies of
service delivery.
13.2 Legislation, Consumer Protection, Privacy

by Design and the IoT
Privacy legislation, current and anticipated, presents significant opportunities for

the advancement of the Internet of Things where stakeholders, including informa-
tion governance practitioners, come together in a new form of strategic partner-
ship to proactively design products and services that have proportionate privacy
considerations built in. This can be achieved by placing the “Privacy by Design”
principle at the heart of product and service design [14–17]. Privacy by design
requires that responses to data protection [protecting privacy] are embedded within
the entire life cycle of a technology and/or service that utilises personal data in
any form, from the start of the design phase, deployment, use recycling and/or dis-
posal. Understanding and applying the philosophy and provisions of privacy legis-
lation as an integral element of product and service design can provide a pathway
for success as people are happier to engage with the IoT when they have more con-
trol over how their personal data will be used, and more certainty that the organi-
sations who process their personal data are capable of protecting and maintaining
privacy.
At the time of writing, the relevant legal framework that informs our thinking
on service design and delivery to assess privacy issues arising from the use of per-
sonal data with the IoT is Directive 95/46/EC (EU Privacy Directive). Directive
2002/58/EC as amended by Directive 2009/136/EC (EU Privacy and Electronic
Communications Directive). Article 5 of the latter requires that public communi-
cations providers such as Internet Service Providers and telecoms companies are
required to take technical and organisational measures to:
ensure the confidentiality of communications and the related traffic data by means of a
public communications network and publicly available electronic communications
services. [18]
Article 6 requires that providers of Web services that transfer messages from
Web servers to Web browsers via text files (cookies) must inform users that these
are being used, describe their use and secure consent before a cookie can be stored
on a user’s device. As those legislative provisions play a lesser role in privacy pro-
tection in comparison with the EU Privacy Directive, assessment of the legisla-
tive frameworks to protect privacy in the development and use of the IoT focuses
on the EU Privacy Directive, and the forthcoming Regulation of the European
Parliament and of the Council on the protection of individuals with regard to the
processing of personal data and on the free movement of such data (EU Data
Protection Regulation) [19].
Naturally, these provisions apply to EU Member States. However, their reach
is not restricted to the geographical boundaries of the EU and IoT device manu-
facturers based outwith the EU may be surprised to learn that they will fall within
the scope of the directive where their devices are used within the EU to process
personal data. Thus, a US manufacturer who produces a pedometer, which trans-
mits data relating to the device owner to their social media feed will, when the
device is used within the territory of a EU member state, fall within the scope of
the legislation.
The sphere of influence of EU Data Protection legislation on the future devel-
opment and operation of the IoT will be substantial, as the EU Data Protection
Regulation provides specific provision in Article 23 that:
[Privacy by design] give incentives to [data] controllers [organisations that decide how an
individuals’ personal information are to be used] to invest, from the start, in getting data
protection right (such as data protection impact assessments, data protection by design
and data protection by default). The proposals place clear responsibility and accountabil-
ity on those processing personal data, throughout the information life cycle. [20]
The effects of the above will be far reaching, as for the first time, legislation
(supported by significant monetary penalties) will require that [legal] entities who
collect and determine the purposes for which personal data will be used, must pro-
actively respond to EU privacy legislation by adopting data protection by design.1
1Also referred to within the legislation as “Privacy by Design.”.

13.2.1 Privacy by Design
Privacy by design first emerged as a term in 1995, from the joint work of
Information and Privacy Commissioner of Ontario, Canada and the Dutch Data
Protection Authority. That work was closely related to “privacy enhancing tech-
nologies” and the principles of “data minimization” [21]:
explored a new approach to privacy protection, with a number of case studies to show that
systems with no personal data—or at least with much less personal data—could have the
same functionalities.
Work to develop the concept of privacy by design continued, culminating in

2009 with the publication of a statement of seven foundation principles [22]. The
third principle “Privacy Embedded into Design” demonstrates how the concept is
an approach of systems engineering, where privacy requirements are considered
and addressed throughout the whole of the engineering process:
Privacy is embedded into the design and architecture of IT systems and business practices.
It is not bolted on as an add-on, after the fact. The result is that it becomes an essen-
tial component of the core functionality being delivered. Privacy is integral to the system,
without diminishing functionality. [22]
While the EU Data Protection Regulation establishes a legislative requirement

for organisations to respond to privacy considerations by adopting data protec-
tion by design and by default, responding to Regulation is not the only driver for
change. The growing number of security breaches, where systems and data are
compromised, requires a fundamental response, at the systems level:
In fact, security breaches may well be a structural problem for an information society that
is increasingly dependent on the good performance of ICT. This should therefore also be
seen as an opportunity for ‘Privacy by Design’. [21]
Given its scope and reach, EU Privacy legislation will be a dominant fac-
tor in the shaping and development of the IoT, not least should privacy by design
become a persuasive, proactive response to privacy protection.
13.2.2 Consumer Trust, Data Sharing Sensitivities

and Market Success
Data, notably personal data, is a critical asset, fundamental to driving technology

innovation. The use of such data, to develop existing applications, services, prod-
ucts and business processes in addition to the creation of completely new ones, are
recognised as a substantial source of economic growth. However, in parallel, con-
sumers’ concerns as to whether organisations can be trusted to make use of their
personal data properly are growing [23]. These concerns are not limited to organi-
sations which use personal data to drive product development. Many organisations
also derive competitive advantage from [big] data mining and analysis that deliver
insights that are used to drive business/organisational decisions sourced from user
behaviour data. Consumers are increasingly concerned about how their personal
data is used in this space [24].
Treacy and Breuning [25] in their assessment of the interfaces between the
development of the IoT and the Data Protection Directive, conclude that for the
IoT to reach its full potential, businesses need to take cognisance of consumer pri-
vacy concerns, putting in place effective strategies to produce products and ser-
vices that consumers are happy to use, with their personal data.
Organisations wishing to take their products and services to the next level [the IoT] will
need to identify the privacy risks and work to mitigate these before embarking on such
projects. [25]
The Article 29 Working Party is comprised of representatives from each EU

Member State’s Data Protection Authority,2 the European Commission and the
European Institutions and publishes opinions and recommendations on elements
of European data privacy laws that it feels are significant [26]. The Working Party
also holds the view that for the IoT to be [commercially] successful, that organisa-
tions must address consumer privacy concerns.
Indeed, empowering individuals by keeping them informed, free and safe is the key to
support trust and innovation, hence to success on these markets. The Working Party firmly
believes that stakeholders meeting such expectations will hold an exceptionally strong
competitive advantage over other players whose business models rely on keeping their
customers unaware of the extent to which their data is processed and shared and on lock-
ing them into their ecosystems. [27]
Consumer and Regulator privacy concerns surrounding the development of the

IoT are understandable, as without the ability to capture, record, transmit, ana-
lyse and further process individuals’ personal data, development of the IoT would
come to a halt.
13.2.3 Personal Data at the Heart of the IoT
The concept of the IoT has been stated as [28]:

A global infrastructure for the information society, enabling advanced services by inter-
connecting (physical and virtual) things based on existing and evolving interoperable
information and communication technologies.
“Things” in the context of the IoT are commonly accepted as smart devices or
objects designed to process data linked to other similar objects or people. That
data typically captures or can be linked to usage activities, which is then often
recorded and/or transmitted. Devices and data are also often associated with
unique identifiers that allow for interaction with other devices/systems within a
2The Information Commissioner’s Office in the case of the UK.

networked environment. It is thus inescapable that data is one of the fundamental

constructs of the IoT, and given the interaction between “things” and individuals,
much of the data are personal data. The Working Party is very clear on this point.
IoT stakeholders aim at offering new applications and services through the collection and
the further combination of this data about individuals – whether in order to measure the
user’s environment specific data “only”, or to specifically observe and analyse his/her hab-
its. In other words, the IoT usually implies the processing of data that relate to identi-
fied or identifiable natural persons, and therefore qualifies as personal data in the sense of
Article 2 of the EU Data Protection Directive. [27]
13.2.4 How Can Stakeholders Come Together to Ensure

Privacy Protection in the Entire Life Cycle of an IoT
Artefact?
In September 2014, the Working Party issued an opinion on the IoT. This identi-
fied the main privacy risks, within the framework of the Data Protection Directive,
and recommendations for addressing those risks.
The recommendations offer a practical view of what IoT stakeholders should consider
when developing and marketing their products in compliance with not only the current EU
data protection framework, but also taking into account [successor legislation] the upcom-
ing EU General Data Protection Regulation. [29]
Shortly after the working Party Opinion was published in January 2015, the
United States Federal Trade Commission (FTC) published its staff report: Internet
of things—Privacy and security in a connected world [30].
The publication of these papers is welcome. Accepting the relatively short
period of time that has elapsed since their publication, it is understandable that
many IoT stakeholders have not yet had the opportunity to consider the issues and
recommendations put forward. A consensus has yet to emerge on how privacy rules
may be applied in a way that strikes the right balance between encouraging and
stimulating innovation of IoT ecosystems and protecting consumer privacy [31].
The Working Party and the FTC share much common ground in their assess-
ment of the nature of the privacy risks emanating from the IoT, for consumers,
many of whom may have no option other than to interact with that ecosystem.
There is also common ground:
that core privacy principles such as transparency consent and data minimisation should
apply in an IoT ecosystem. [31]
A significant difference between the markets of the United States and European
Union lies in the fact that federal data protection laws only exist in European
Member States. Not only is the Data Protection Directive well established, this
legislative framework will shortly be extended with a single Data Protection
Regulation [19]. Thus, in terms of exploring frameworks within which IoT stake-
holders can purposefully come together to ensure proportionate responses to
consumer privacy concerns throughout the entire life cycle of an IoT device and
consequent processing and transmission activities, the opinion of the Working

Party therefore is a valuable foundation and reference point. As the IoT develops
in parallel with privacy jurisprudence, more specific mature guidance on establish-
ing privacy as a core component of product and service by design will emerge. It
is fundamental that future thinking in this space emanates from a multidisciplinary
focus—technologists and privacy/information governance practitioners need to
come together, otherwise innovation will become stifled.
The IoT will not just requite technological innovation. Legal innovation will be at a premium.
New thinking and new paradigms are required if IoT stakeholders, many of whom are based
in the US, are to have any hope of complying with perspective and evolving EU privacy laws
[and increased customer demands for privacy]. One internet, one thing, two worlds. [32]
13.2.5 A Roadmap for Collaboration
In framing IoT privacy considerations, the Working Party started with a pragmatic
baseline, working within the scope of real-world IoT facets which are currently
commercially viable: wearable computing; the quantified self and home automation
or domotics [27]. While the future direction of IoT developments is uncertain, as
other possibilities for viable commercial applications have yet to emerge from the:
convergence and synergies of the IoT, with other technological developments such as
cloud computing predictive analytics [27].
The European Commission are committed to: “implementing a single, tech-

nologically neutral and future-proof set of rules across the EU” [32]. Given the
jurisprudence of data protection laws in the UK and Europe since the Younger
Committee on Privacy [33], the Council of Europe Convention 1981 [34] and the
EU Commission proposal for reform of the Data Protection Directive are derived
from and continue to be focused on principles that can be applied to all forms
of electronic processing of data—the recommendations from the Working Party
Opinion [27] are likely to remain relevant, capable of being applied to cover future
IoT developments, as that ecosystem evolves.
13.2.6 The Extent to Which IoT Stakeholders Are Engaged

by the Legislation: Roles and Responsibilities
To develop effective responses to IoT privacy concerns, an important step is to

understand the roles and responsibilities that IoT stakeholders have in relation to
the legislation. There are two primary actors, the data subject, who is a (living)
person that is the subject/focus of the personal data and the data controller, who is
a legal entity that determines either alone or in concert with another party (a joint
data controller) the purposes for which personal data will be used. The legislation
only applies to the data controller.
In 1999, the [then] chief executive officer of Sun Microsystems, Scott

McNealy, stated to a group of reporters and industry analysts that consumer pri-
vacy issues are a “red herring” and that “You have zero privacy anyway… Get over
it.” [35]. McNealy’s statement failed to take into account, at least in the European
context, the fact that where an IoT stakeholder qualifies as a data controller, they
have significant responsibilities for protecting and maintaining the privacy of cus-
tomer or data subjects’ personal data. From a consumer and legislative perspective,
data controllers cannot afford to bring to market devices and/or services that are
not capable of maintaining customer privacy. Arguably therefore, privacy is a core
functional design element of IoT things and services. Responsibility for protecting
the privacy of the consumer starts and ends with the data controller.
The Working Group in establishing their view of the IoT identified six key
stakeholders. Table 13.1 then sets out the relevant stakeholder roles.
Table 13.1 Stakeholder responsibilities

Stakeholder role Notes
Device Defining the functionality of a device and creating the ability for it to operate
manufacturers means that a device manufacture will determine what data is captured and
the subsequent modes of processing/operation, which can include onward
transmission of data to another device or service provider. Determining the
purpose of data processing qualifies a device manufacture as a data controller
Social platforms Data subjects may share their personal data, captured via a range of devices
via social media. Sharing of data collected and aggregated by IoT “things”
on social networks typically happens automatically via default settings con-
figured by the user. Personal data pushed to a social media platform will be
processed by the service provider for distinct purposes, established by that
provider. This will then qualify the provider as a data controller
Third-party App developers process personal data via APIs. Unless the data received/
application collected by the API for processing has first been anonymised, the app devel-
developers oper will have determined the purposes for data processing and will qualify
as a data controller. The app provider must clearly inform the user as to how
their personal data will be processed. Otherwise, informed consent will not
have been provided and continued processing will be unlawful
Other third A third party could take the form of an insurer, who provides pedometers
parties to monitor exercise, with the aim of adjusting health insurance premiums
accordingly. The third party, unlike the device manufacture has no control
over what data is collected by the device. The insurer has determined that the
physical activity of a person will be measured in order to offer lower insur-
ance premiums. Determining that purpose of data processing qualifies the
insurer as a data controller
IoT data Cloud providers that store data collected through IoT things will be data
platforms controllers, as they determine how data will be stored, secured, received and
transmitted between devices, etc., thus qualifying that service provider as a
data controller
Individuals as Users of IoT devices can qualify as data controllers where they collect and
data subjects: process the personal data of others, for non-domestic purposes. The use of
subscribers, smart glasses is likely to collect personal data about others
users, non-users
13.2.7 Legal Basis for Processing Personal Data as a

Foundation for Product and/or Service Design
In responding to privacy as a core functional design element of IoT things and

services, understanding the legal basis for processing personal data is fundamen-
tal. Before personal data can be used, IoT stakeholders who provide devices and/
or services (where the provider qualifies as a data controller) must ensure that
their devices/services are capable of fulfilling at least one of the six requirements
of Article 7 of the EU Data Protection Directive. That said, in reality only three
requirements are likely to be of concern to product designers and service deliver-
ers. Critically, products and services must be designed and managed so that they
are capable of successfully engaging those requirements.
This will by no means be an easy task: providing users with the functionality
to have full control via consent on how their personal data is used and evidencing
that consent was or was not provided (potentially continually opting in and out of
services and device functions) may require that granularity at the heart of prod-
uct and service design becomes a fundamental consideration. The three require-
ments of the six available from Article 7, that are more likely to apply as set out in
Table 13.2.
Table 13.2 Requirements of EU data protection directive

Requirement Notes
Consent People need to be fully informed as to how their personal data will be used,
and by whom; where a user opts to consent, that consent must be explicitly
captured and that fact recorded; users also have the right to withdraw their
consent—this will have to be managed as part of product and service design.
Fundamentally, IoT systems design must provide for robust consent manage-
ment—where users can continually opt in or out, without any disadvantage:
they must retain the right to have full use of the functionality of the system/
service that they have paid for [36]
Contract Use of personal data by IoT devices and/or services can be legitimised where
there is a contract between the data controller and the data subject. The use of
personal data must be necessary to fulfil the contract, requiring “a direct and
objective link between the processing itself and the purposes of the contractual
performance expected from the data subject” [27]. For the contract to remain
valid, there cannot be any creep in the use of personal data. The collection and
use of personal data must be clearly understood and defined as part of device
and/or service design
Legitimate A data controller can process personal data, and share this with a third party
interests where it is their legitimate interests or those of the third party as long as the
fundamental rights of the data subject are not undermined. As the privacy
concerns of data subjects are fundamental, it is unlikely that a data controller
can successfully claim economic interests as a justification to legitimise their
processing of personal data
13.2.8 Challenges: Protecting Privacy Within the Internet

of Things
The Working Party in forming a view of the privacy challenges related to the IoT
identified six areas of concern as set out in Table 13.3. The Working Party’s views
on how these challenges can be addressed, as design considerations for the IoT are
set out in Tables 13.4 and 13.5.
13.2.9 IoT Privacy Design Requirements
The Working Party’s recommendations provide guidance to IoT stakehold-

ers on the design and service considerations that in their opinion require to
be addressed to establish an IoT environment that is compliant with privacy
Table 13.3 Areas of concern

Privacy challenge Impact
Lack of control and Given the ubiquitous nature of the IoT, where a stakeholder pro-
information asymmetry cesses personal data unknown to the user, people may find that they
rapidly lose control of their privacy, where they then become subject
to third-party monitoring, notably where their personal data is dis-
seminated to other stakeholders without prior knowledge or consent
Quality of user’s consent It is relatively easy for IoT stakeholders to be invisible to users. If a
user is unaware of the data processing that is taking place, then con-
sent cannot be relied upon as a lawful basis for processing personal
data. Data subjects must be informed that processing is taking place
Inferences derived from With the increased volume of data generated by the IoT, combined
data and repurposing of with advances in data analysis and cross-matching, it becomes
original processing easier for secondary forms of personal data to be generated and
used for purposes beyond those that were originally intended
Intrusively bringing out In the IoT the proliferation of sensors/devices, makes it relatively
of behaviour patterns and easy to build up a picture of a person’s life from trivial or even
profiling anonymous data. Data harvested from the IoT, could be used to
predict future behaviours, leading to significant privacy intrusion,
where a data controller makes a decision on an individual, based on
future profiling
Limitations on the The nature of the IoT is likely to make it extremely difficult for
possibility to remain users to use services anonymously, as the connection between a user
anonymous when using and a device will more often than not be inextricable
services
Security risks: security It may be difficult to implement many security measures on
versus efficiency IoT devices such as sensors, where there is a trade-off between
hardware-based encryption and battery life. Integration of physical
and logical IoT components, provided by a range of stakeholders
only provides a level of security at the weakest point in the chain.
IoT devices that become everyday objects present a new distributed
target
Table 13.4 Privacy design requirements for all IoT stakeholders

Requirement Action
Privacy impact PIAs undertaken prior to the launch of any IoT entity.
assessments PIA methodology recommended for RFID applications [37] should be
(“PIAs”) considered
User empowerment Data subjects rights must be recognised and respected—users must
retain control over their data at all times
Data subjects as consumers/users should not suffer any economic penalty
or service degradation if they opt not to consent to the use of their
personal data. Consent should be granular—focused on specific areas of
processing. Data subjects should have the facility to continually with-
draw their consent, without having to exit from the service provided [38]
All IoT stakeholders must be able to communicate to ensure that user
choices are respected and acted upon. IoT devices and services should
operate with a do not disturb function—including the facility to disable
and enable sensors
Data minimisation Most IoT stakeholders only require aggregated data
Stakeholders should delete raw data as soon as that data has been
extracted for processing
Deletion should take place at the closest point of data collection of the
raw data
Privacy by design Principles of privacy by design and privacy by default to be applied by
and privacy by all IoT stakeholders
default
Transparency Information on the use of personal data by IoT stakeholders should be
made available in as user-friendly a manner as possible. Such informa-
tion should not be confined to general privacy statements that are avail-
able from terms and conditions
legislation. Recommendations that are common to all IoT stakeholders are set out
in Table 13.4. An overview of the obligations that are specific to a stakeholder are
provided in Table 13.5.
Thus, those who seek to take their products and/or services into the IoT eco-
system will need to understand the fundamental concepts of privacy legislation,
and work to mitigate privacy concerns as a core element of product and/or service
design. A clearer picture is beginning to emerge of the initial suite of design con-
siderations that IoT stakeholders will have to address.
13.2.10 Starting Points for Multidisciplinary Working

to Innovate Across Technology and Privacy
Legislation
Considering the recommendations and their context (Tables 13.4 and 13.5) rein-
forces the earlier point that development of the IoT will require multidiscipli-
nary working as a level of technical and legal innovation will be a prerequisite
for success. Therefore, a fundamental consideration is how best can the relevant
Table 13.5 Privacy design requirements for specific IoT stakeholders
216
OS and device Application developers Social platforms IoT device holders and Standardisation bodies and
manufacturers additional recipients data platforms
Privacy impact Special attention to types of
assessments (“PIAs”) data being processed—nota-
bly the possibility of infer-
ring sensitive personal data
Privacy by design Dedicate components Minimise personal data col- Develop lightweight encryp-
and privacy by to provide encryption lection—strictly to purposes tion and communication
default services of processing protocols
Control by the user Provide users with tools Data formats clear and self-
to read, edit and modify explanatory: users should
data before this is made be clear on what personal
available to a data data is being collected and
controller transferred
Transparency Users of IoT devices should
have the functionality to
inform others that their per-
sonal data may be collected
in the presence of a device
Data Portability Users should be enabled Tools should be provided Default settings prompt Promotion of portability and
to transfer their data to for users to export both raw users to review, edit and interoperability standards
other devices and/or aggregated data in a confirm before data from
standard and usable format IoT device is published to
social media
Data minimisation Minimise storage of per- Minimise the amount of Focus on the format for raw
sonal data on device to personal data collected to and aggregated data.
that which is necessary that required to provide the Promotion of data formats
Raw data transferred to service that contain minimal numbers
aggregated data on the Users should be offered of strong personal identifiers
device the facility to use services to facilitate anonymization of
Provide the facility anonymously IoT data
to irreversibly delete/
destroy personal data
C. Milne and S. Watt
multidisciplinary elements come together in academic, professional, practitioner

and organisational contexts to provide and sustain the required technical and legal
innovation necessary to provide for an effective IoT ecosystem?
Understanding and recognising when data makes the transition from data to
personal data or sensitive personal data will be critical in designing and provid-
ing effective privacy solutions for the IoT. In that regard, an area where technical
and legal innovation can come together is the area of developing and integrating
anonymization techniques [38] into the design process. The purpose of anonymi-
zation is to turn data into a form which does not identify individuals and where
identification is not likely to take place. This will allow for a much wider use of
the information, while mitigating privacy risks for the data subjects. Successfully
anonymized data will also fall out of the scope of data protection legislation,
which by extension will reduce pressures on IoT stakeholders where the scope of
their responsibilities as a data controller can be reduced.
Working to embed Privacy by Design principles as a foundation of systems
analysis and design will also be a significant step. This may also involve estab-
lishing new interfaces/partnerships (see earlier theme of the chapter), with design-
ers, engineers, information governance and privacy practitioners coming together
within the product design and requirements specification phases. For example,
the requirement to advise users as to how their personal data is being processed,
which will include how data is collected and transferred across the IoT, to other
stakeholders who are data controllers, could be aided by repurposing business
data flows/process maps to explain what processing takes place and when to users.
This level of analysis can also be utilised, in an attempt to reduce the likelihood of
incremental creep in the processing of personal data.
Mapping out the processing and understanding data flows, notably when
data can take on new meaning, will assist in establishing processing boundaries.
Having determined the process boundaries, these should then inform the design
phase, with the view of developing products/services that guard against any drift
into illegal uses of personal data, notably where inferences can be derived from
data and repurposing beyond the lawful justification for processing. This may
mean that data controllers can rely more heavily upon contract as a legitimising
basis to process personal data, reducing the requirement to rely on consent—
which could then simplify product and service design.
13.3 Conclusions
To respond successfully to the changing landscape, creating the capacity to secure

new possibilities notably where ICT services, data analytics and a maturing and
consumer-friendly IoT provides a higher degree of business and societal transfor-
mation, a number of fundamental changes to product design and service delivery
need to be considered and made.
The traditional approach to delivering enterprise ICT is being replaced by an

approach which recognises that more fulfilling partnerships are a prerequisite if
we are to succeed in meeting organisational and customer demands. Cloud com-
puting has provided one such opportunity which allows an ecosystem to be created
which spans traditional boundaries—shifting organisational mindsets that others
can deliver some core functions more efficiently and effectively than in-house;
thus freeing up time and energy to add value elsewhere.
However, to capitalise on this changes to working practices, resource and con-
tract management are required. The IoT provides a useful focus on crystallising
the mechatronics challenges of the future as we see these. Consumers and legisla-
tors require that all IoT stakeholders recognise when personal and sensitive per-
sonal data are being processed, and that their products and services are capable
of protecting privacy by design. The IoT will collapse, where consumers do not
release their personal and/or sensitive data to device manufacturers and service
providers.
The immediate (0–5 years) challenges are therefore:
• Recognising the need to form multidisciplinary partnerships—opportunities to
innovate in product design and service delivery will be lost where there is no
environment for technologists and information governance and privacy practi-
tioners to come together. Organisations that establish that form of partnership
working will be more likely to secure a competitive advantage;
• Establishing new methods of securing and managing user consent—ensuring no
economic or functional disadvantage if a user opts out of data sharing;
• Recognising what is and is not personal data in the IoT as processing extends
and diversifies in that environment;
• Embedding and responding to privacy by design as a fundamental construct of
product and service design.
The medium term (5–10 years) challenges are likely to be:
• Recognising what is and is not personal data in the IoT as processing extends
and diversifies in that environment;
• Maintaining a regulatory regime and jurisprudence that protects privacy, with-
out undermining the capacity to innovate;
• Understanding how best to educate and inform technologists, legal/juris-
prudence practitioners so that legal and societal demands are capable of
being addressed in the development, implementation and application of new
technologies.
The long term (10–20 years) challenges are likely to be:
• Understanding how best to educate and inform technologists, legal/juris-
prudence practitioners so that legal and societal demands are capable of
being addressed in the development, implementation and application of new
technologies;
• Quantum computing—to what extent will this alter our perceptions on privacy
and technological innovation?
References
1. Handy C (1993a) Understanding organizations, 4th ed, Penguin, p 366

2. Handy C (1993b) Understanding organizations, 4th ed, Penguin, p 257
3. Prahalad CK, Krishnan MS (2008) The new age of innovation: driving co-created value
through global networks. McGraw-Hill, USA, pp 22–23
4. Jerry Niman IT Services Ltd (2014) UK HE cloud computing implementations @ jerryni-
man.co.uk/core_modules/entitylist.php?Entity=Cloud_Collaboration_Implementations.
Accessed 13 Nov 2015
5. Ahmed M (2015) HP says it overpaid by $5bn for autonomy, financial times @ http://www.
ft.com/cms/s/0/35db8c1e-ef2d-11e4–87dc-00144feab7de.html#axzz3uNpFyqf8. Accessed 31
July 2015
6. Bishop T (2015) Satya Nadella reveals Microsoft’s new mission statement, sees ‘tough
choices’ ahead, Geekwire @ www.geekwire.com/2015/exclusive-satya-nadella-reveals-
microsofts-new-mission-statement-sees-more-tough-choices-ahead/. Accessed 31 July 2015
7. World Economic Forum Global Risks Report (2015) @ www.weforum.org/reports/global-
risks-report-2015. Accessed 12 Oct 2015
8. ISACA (2012) Governance of enterprise IT (GEIT) Survey @ www.isaca.org/Pages/2012-
Governance-of-Enterprise-IT-GEIT-Survey.aspx. Accessed 12 Oct 2015
9. Hyung Jun K (2012) Security and vulnerability of SCADA systems over IP-based wireless
sensor networks. Intl J Distrib Sens Netw, Article ID 268478
10. Bradley DA, Russell D, Ferguson I, Isaacs J, White R (2015) The internet of things—the
future or the end of mechatronics. Mechatronics 27:57–74
11. Rodrigo R, Najera P, Lopez J (2011) Securing the internet of things. IEEE Comput Soc,
51–58
12. Prahalad CK, Krishnan MN (2008) The new age of innovation: driving co-created value
through global networks. McGraw-Hill, USA
technology. Harvard Business School Press, USA
14. Schaar P (2010) Privacy by design. Identity Inf Soc 3(2):267–274
15. Cavoukian A, Taylor S, Abrams ME (2010) Privacy by design: essential for organizational
accountability and strong business practices. Identity Inf Soc 3(2):405–413
16. Radomirovic S (2010) Towards a Model for Security and Privacy in the Internet of Things,
1st Intl. Workshop on Security of the Internet of Things, @ www.nics.uma.es/pub/seciot10/
files/pdf/radomirovic_seciot10_paper.pdf. Accessed 12 Oct 2015
17. Weber RH (2010) Internet of things—new security and privacy challenges. Comp Law Secur
Rev 26:23–30
18. Directive 2002/58/Ec of the European Parliament and of the Council of 12 July 2002 @
http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32002L0058&from=EN.
Accessed 8 Aug 2015
19. Proposal for a Regulation of the European Parliament and of the Council on the protection of
individuals with regard to the processing of personal data and on the free movement of such
data (General Data Protection Regulation), COM (2012) 11 final @ http://ec.europa.eu/jus-
tice/data-protection/document/review2012/com_2012_11_en.pdf. Accessed 8 Aug 2015
20. Article 29 Data Protection Working Party (2012) Opinion 01/2012 on the data protection
reform proposals @ http://ec.europa.eu/justice/data-protection/article-29/documentation/
opinion-recommendation/files/2012/wp191_en.pdf. Accessed 9 Aug 2015
21. Hustinix P (2010) Privacy by design: delivering the promises. Identity Inf Soc 3(2):253–255
22. Privacy by Design @ https://www.privacybydesign.ca/index.php/about-pbd/7-foundational-
principles/. Accessed 10 Oct 2015
23. Roeber B, Rehse O, Knorrek R, Thomsen B (2015) Personal data: how context shapes con-
sumers’ data sharing with organizations from various sectors. Electron Markets 25(2):95–108
24. Zhang Xiaoni, Xiang Shang (2015) Data Quality. Analytics, and privacy in big data, big data
in complex systems—studies in big data 9:393–418
25. Treacy B, Breuning P (2013) The internet of things: already in a home near you. Priv Data
Prot 14(2):11–13
26. Carey P (2009) Data protection: a practical guide to UK and EU law. Oxford University
Press, UK
27. Article 29 Data Protection Working Party (2014), Opinion 8/2014 on the on Recent
Developments on the Internet of Things @ ec.europa.eu/justice/data-protection/article-29/
documentation/opinion-recommendation/files/2014/wp223_en.pdf. Accessed 9 Aug 2015
28. Internet of things—global standards initiative @ www.itu.int/en/ITU-T/gsi/iot/Pages/
default.aspx. Accessed 21 Dec 2015
29. Salgado M (2014) Internet of things revisited. Priv Data Prot 15(1):12–14
30. FTC Staff Report (2013) Internet of things: privacy and security in a connected world @
www.ftc.gov/system/files/documents/reports/federal-trade-commission-staff-report-novem-
ber-2013-workshop-entitled-internet-things-privacy/150127iotrpt.pdf. Accessed 10 Oct 2015
31. Corbet R (2014) Internet of things where are we now? Priv Data Prot 15(6):13–16
32. European Commission (c2011) How will the EU’s reform adapt data protection rules to
new technological developments? @ http://ec.europa.eu/justice/data-protection/document/
review2012/factsheets/8_en.pdf. Accessed 10 Oct 2015
33. Cmnd 5012, Sir Kenneth Younger (Chairman) (1972) The report of the committee on privacy,
HMSO
34. Council of Europe (1981), Convention for the protection of individuals with regard to auto-
matic processing of personal data @ http://www.coe.int/en/web/conventions/full-list/-/con-
ventions/treaty/108. Accessed 10 Oct 2015
35. Sprenger P (1999) Sun on privacy: ‘get over it’ wired @ archive.wired.com/politics/law/
news/1999/01/17538. Accessed 10 Oct 2015
36. Article 29 Data Protection Working Party (2011) Opinion 13/2011 on geolocation services
on smart mobile devices @ ec.europa.eu/justice/policies/privacy/docs/wpdocs/2011/wp185_
en.pdf. Accessed 15 Nov 2015
37. Article 29 Data Protection Working Party (2011) Opinion 9/2011 on the revised industry pro-
posal for a privacy and data protection impact assessment framework for RFID applications
@ ec.europa.eu/justice/policies/privacy/docs/wpdocs/2011/wp180_en.pdf. Accessed 15 Nov
2015
38. Article 29 Data Protection Working Party (2014) Opinion 05/2014 on anonymisation
Techniques @ http://ec.europa.eu/justice/data-protection/article-29/documentation/opinion-
recommendation/files/2014/wp216_en.pdf. Accessed 15 Nov 2015
Chapter 14
Engineering Design for Mechatronics—
A Pedagogical Perspective
Simeon Keates
14.1 Introduction
Since the introduction of undergraduate and postgraduate programmes in

mechatronics and related subjects in the mid-1980s, there has been a near continu-
ous debate as to the nature and standing of mechatronics both as an Engineering
discipline and in relation to its role within Engineering Design [1–5].
In the case of mechatronics education, what has emerged is a wide variety
and range of courses structured around the basic tenets of integration concen-
trated around the core disciplines of Electronics, Mechanical Engineering and
Information Systems or Computing but with a wide range of variation and variety
to accommodate local requirements and conditions.
Thus, a course developed and delivered in, say, Detroit [6], is likely to differ
significantly from one in place in Singapore [7], while both have entirely legiti-
mate claims and arguments to be considered as mechatronics programmes.
Notwithstanding this difference in emphasis, each course will, in general, seek
to conform to the requirements of achieving an appropriate level of integration
between the core disciplines, with an emphasis appropriate to the overall require-
ments of the course.
Here, we examine how innovative and challenging mechatronics programmes
structured to meet future needs must still incorporate the basic principles of
Engineering Design. However, mechatronics remains a fundamentally innova-
tive field and simple instruction in the basic mechanics of putting the components
together is missing an educational opportunity to push students to develop their
creative engineering thinking. Mechatronics, being such a diverse field, allows
students and teachers to explore genuinely innovative questions and solutions. As
such, it is well suited to allowing teachers to set tasks and projects for students
S. Keates (*)
Faculty of Engineering and Science, University of Greenwich, Chatham Maritime, UK

DOI 10.1007/978-3-319-32156-1_14
222 S. Keates
that break new ground and explicitly support the creation of the new concepts and
solutions required to take mechatronics forward.
When looking at mechatronics-oriented degree programmes it is necessary to
consider how mechatronics is likely to develop and change in the mid- to longer-
term future. The goal of any good degree programme is to not only prepare each
student to secure their first job, but also to give them the correct skills and mind-
sets to retain employment throughout their entire working life. This goal is a par-
ticular challenge in a discipline that is as diverse as mechatronics.
14.2 Learning Objectives of Mechatronics Courses
As the name mechatronics implies, the subject is generally considered to be a

merger of both traditional Mechanical and Electrical/Electronic Engineering, often
with computing elements. However, while knowledge of both engineering disci-
plines allows students to understand how mechatronic systems function, it is sug-
gested that an essential component of any mechatronics programme is Engineering
Design. Mechatronics students are not typically driven solely by grades, although
this is an undeniably important motivational factor for the brightest students in par-
ticular. Instead, most mechatronics students are more generally motivated by the
desire to solve a problem. Any educational programme should be oriented to sup-
port this desire and must not inhibit it through too much formulisation. In other
words, mechatronics programmes need to support open-ended active enquiry rather
than do-it-yourself flat-pack or pro forma type assembly instructions. It is proposed
that the key attributes of a graduate of a mechatronics programme are:
• Confidence • Skills
• Creativity • An ability to work in a team
Figure 14.1 [5] shows that Engineering Design can be placed at the intersection
of a science-based set of skills, the horizontal element of the figure, and social and
artistic skills, the vertical element. To these must be added a wider awareness of
a range of issues necessary to convert a concept into a viable system or product,
such as aesthetics, manufacture, ergonomics and human factors.
In considering the requirements of a mechatronics course with Engineering
Design at its core, the essence remains that of balancing the Engineering and IT
content within a design focus that supports both individual and group working.
The latter is especially important for mechatronics, which is a confluence of very
diverse technical domains and thus any one person is unlikely to be a master of
all of the technical skills required to build a successful device or system, par-
ticularly within the context of developments such as cyber-physical systems and
the Internet of Things. In industry, most graduates will be expected to work in a
team and so ought to experience the realities of such co-operative work in their
programmes.
14 Engineering Design for Mechatronics … 223
Fig. 14.1 Engineering design issues (after [5])
Theoretician Pragmatist
Reviews design process and Emphasis on problem solving
outcomes Selects, uses and refines
Identifies methods and methods
generates procedures Generates solutions
Establishes good practice Establishes practice
Refines theories
Fig. 14.2 Approaches to design
Key elements here are the need to support communication between members of
the group, for instance through computer-based communications structured around
the use of digital libraries [8, 9], and to expose students, both individually and as
members of a group, to the design process from concept development to imple-
mentation [10]. Intrinsic to this is the need to ensure that, particularly in a cross-
and interdisciplinary environment, issues of potential misunderstanding through
different and differing use of terminology is avoided [11].
Further, it has been suggested [5] that design can be categorised in relation to
two broad approaches; theoretical and pragmatic, as illustrated by Fig. 14.2. In
practice, these extremes do not exist in isolation, but co-exist along a continuum
within the design process. What is perhaps of more significance in relation to
course design is that students, inevitably, lack the range of experience associated
with established design engineers, and this then impacts on their approach to prob-
lem solving [12, 13].
Here, we shall consider issues associated with achieving a design-based
input through a combination of project- and problem-based learning linked to
mechatronics and looks at these from a range of perspectives including the need to
encourage innovation and student perception [14–19].
224 S. Keates
14.3 The Challenge of Teaching “Innovation”
Innovation and, by extension, the ability to innovate, is a key element of any

Engineering Design process and one that needs to be encouraged and developed
within a mechatronics course. In the widest sense, the ability to innovate impacts
upon issues such as market penetration and the ability to develop, implement and
introduce new products to market ahead of competitors, and to maintain that posi-
tion over time.
Typically, innovation is seen as a continuous and dynamic process involving
investigation and feedback across a number of individuals. However, until rela-
tively recently, innovation was considered by many companies as a closed process.
An alternative approach, that of open innovation, takes as its goal not simply pre-
serving a current market, but actively seeking to grow and develop other market
areas through importing ideas, concepts and technologies as appropriate.
14.3.1 Open and Closed Innovation
Innovation, in all its potential forms, is key to the achievement of new generations
of products and systems. In order to develop and take forward the innovative pro-
cess to meet a new set of challenges, Chesbrough [20, 21] has suggested the need
for a shift from the traditional approach, defined as Closed Innovation, with its
orientation towards secrecy and the retention of ideas to one of Open Innovation
in which ideas and solutions are widely sought from both within and from outside
the organisation.
The relationships between these two divergent approaches can be seen in
Figs. 14.3 and 14.4. From these, it can be seen that they each represent a signifi-
cantly different focus on the innovation process, both in terms of the value of ideas
Fig. 14.3 Closed innovation

Fig. 14.4 Open innovation
and the ways in which such ideas are to be incorporated into that process. The
revised methodology represented by open innovation has been adopted by organi-
sations such as Proctor & Gamble [22] and the US Department of Education [23]
to create platforms to develop and take forward new ideas, but perhaps more
importantly to bring in new ways of thinking from outside the organisation.
Similarly, IBM runs an annual “Innovation Jam” as part of its Global Innovation
Outlook [24]. Though the underlying motivation, in one case growing company
profitability and in the other enhancing an education system, may differ, both are
exhibiting a degree of openness by inviting external bodies, groups and individuals
to submit their ideas into a central ‘pot’ for consideration.
14.3.2 Students and Innovation
In 1998, John Prados [25] suggested that Engineering graduates were perceived as
having a range of weaknesses, including:
• Technical arrogance
• Lack of design capability or creativity
• Lack of appreciation for considering alternatives
• Lack of appreciation for variation
• Poor overall perception of the project
• Narrow view of engineering and related disciplines
226 S. Keates
Fig. 14.5 Design support tools
• Weak communication skills

• Little skill or experience in working in teams
In developing innovative thinking by students, all the above issues need to be con-
sidered, some of which may well, however, be in conflict with the administrative
requirements associated with grading and the ability to differentiate between indi-
vidual students in assessment schemes [26–32].
There is a range of tools available to support both the design process (Fig. 14.5)
and communications between members of the design group [8, 9, 33,34]. In terms
of encouraging an innovative approach to design problems, in which the aim is
encouraging students to bring forward new and novel ideas, there is a need to cre-
ate an environment where trying and failing is not considered as a failure in rela-
tion to a student’s ability to progress or pass the course or module. This means that
students are then free to put forward ideas and pursue options in an environment in
which the emphasis is on trying and not on failing, i.e.: “Try and fail, but don’t fail
to try”.
However, students often focus on the requirements necessary to achieve a par-
ticular grade, which in turn tends to lead them to be conservative in their approach
as they attempt to ensure that they achieve the necessary marks for the target
grade. This conservatism then runs contrary to the requirement to encourage inno-
vation at the expense of an occasional failure to achieve set goals. Thus, insistence
on the allocation of a grade, and of differentiating between students, can have a
negative impact on the level of innovation.
In this respect, consider student reaction to the essay topic: “Eli Witney and the
origins of mass production”, which was posed in a manufacturing course. Students
were told:
• That there was no predefined or predetermined content required to achieve a
particular grade.
• That the emphasis was to be on their ability to source, organise and interpret
data available from a variety of sources.
• That in order to obtain a passing grade they were required to demonstrate that
they had carried out a level of research and analysis associated with basic infor-
mation gathering.
• That to achieve a higher grade they were required to demonstrate that they
could organise and arrange the information to tell a specific story of their choice
using the title as guide.
• The length of the paper.
A comparatively small number of students took advantage of the flexibility to
develop a case while the majority took the conservative approach of ensuring they
did what was required to pass but then did not feel that they wished to take on
what they perceived were the potential risks associated with the achievement of a
higher grade.
14.3.3 Choice of Tools
Once a design brief has been given to students, they are then typically given access
to a workshop or laboratory for construction of their solutions. The equipment
and construction components they are given access to will influence their design
process. For example, it is common to use standard components such as Arduino
boards and associated sensors [35] or Lego Mindstorms [36] in first or second year
mechatronics projects. The choice of which of these components are available
will push students down particular design paths. While such provision may sim-
plify the project for the students, as well as keep costs down, it does come at the
expense of a level of restriction on design creativity.
One possible solution to the cost issue is the use of computer simulations
of components through the kinematic modelling of their properties. An exam-
ple of such an approach was the variable fidelity prototype developed for the
Interactive Robotics Visual Inspection System (IRVIS) [37], which was an accu-
rate model of both the size and kinematic response of robot with five motors
and five degrees of freedom—see Fig. 14.6a, b. Such a virtual prototype can be
reconfigured, redesigned and completely altered with a few lines of code for
absolutely no cost.
The advantages of using a working model that can be adjusted quickly and for
comparatively little resource cost when trying to teach innovation are obvious.
Students are encouraged to explore different options, because the effort involved
228 S. Keates
Fig. 14.6 Virtual
prototyping in design
education, a interactive
robotic visual inspection
system (IRVIS) consisting
of a camera mounted on a
gantry above a moveable
tray of microcircuits. The
robot has 5 degrees of
freedom. b the variable
fidelity prototype—a virtual
model of the IRVIS robot
with authentically modelled
kinematic performance
in creating alternative options is minimal and the feedback on the success or oth-
erwise of their design is very quick. However, the model does need to be flexible
enough to support more radical design solutions, otherwise what may be intended
as a tool to promote innovation may itself become a limitation on that same inno-
vation if students cannot explore and examine all of the design variations they can
conceive.
14.4 Approaches to Assessment
As design is generally a group or team exercise, it is sensible to incorporate a

group design exercise within a design-oriented mechatronics course. This, how-
ever, leads to issues of ensuring that the marks and grades reflect the contribution
of the individual members of the group. Strategies that have been used include:
Flat marks Each member of the group receives the same mark irrespective of their
contribution to the final report. This can work if balanced by the internal peer
pressures of the group ensuring a balanced level of activity across all members
Individual Assessing an individual student’s contribution could typically involve an
contribution agreed introduction and conclusion for which each member of the group
would be awarded a shared mark. The individual contributions to the overall
project would then be identified and the sections of the report associated with
particular responsibility and activity graded separately
This approach generally works best where group members have either
identifiable skills or worked on clearly demarcated components. The classic
example is the development of a robot for following a white line where
students can be allocated responsibility for building (i) the robot chassis;
(ii) the sensor array; and (iii) the control code
Combined An alternative approach is to couple the project work with an examination that
marking is designed to establish a student’s overall depth of knowledge of the project.
For example, students are first asked to write a group project report, which
is then graded for the whole group. The group is then invited to make an
(ungraded) presentation on the report summarising the key findings. The
students are free to decide who presents what. This presentation is then
followed by individual oral exams, where the group project marks can be
increased or decreased by up to one grade
Such an approach gives the students an incentive to work well as a group,
because they all benefit from a high initial report grade. However, the students
feel some degree of confidence that weaker members of the group will be
found out in their individual exams and so there is an element of correction
in the final grade. Similarly, very able and diligent students also have the
opportunity to improve their grade if there had been a problem elsewhere in
the group
Peer Peer assessment can be used in association with either of the above but with
assessment a proportion of the marks being held back to be allocated by members of the
group to the other members of the group to reflect their perceived contribution
Each of the above has been used in association with group projects in design,
and each has been met with various degrees of scepticism by students.
However, the general view was that the overall marks awarded reflected the
contribution by the individual group members
A further approach used where groups were competing on the same project
brief, as for instance representing individual design groups tendering for
a project, was to distribute the reports to other teams prior to marking and
asking for a critique of the these to be submitted. These critiques were then
graded, with the grade then contributed a percentage of the overall grade. The
results from these critiques were generally very interesting, as the majority of
teams did not set out to attempt to destroy the other’s case, but to genuinely
perform a critical analysis of the proposal. Two instances are of particular
interest:
• One group commented that they wished they had thought of an idea put
forward by another group and followed this up with a detailed analysis to
demonstrate why they still thought that their solution was superior
• Another group commented to the effect that after doing the critique
remarked on “the problems of grading such reports” and that they had never
appreciated these previously
230 S. Keates
14.4.1 Measures of Success and Success Criteria
The challenge of how to grade such reports is interesting. In any design activity,
one of the earliest considerations is that of what measures of success are to be
used. Put simply, if two designs are to be compared, what evaluation criteria are to
be used? Again there is a range of possible strategies.
For example, consider the classic Civil Engineering student design problem,
that of building a structure to span a gap supporting a specified weight at the mid-
point. Typical measures of success are (i) whether the structure supported the load;
(ii) the weight of the structure; and (iii) the “cost” of the structure, which is usu-
ally calculated based on the cost of the components and the labour time for fabri-
cation. Most students typically design a traditional truss-type structure, usually a
Pratt or Warren truss, because that is what they automatically assume will be the
most effective structure. In reality a Waddell-type truss, i.e. a very large triangle
design is usually the most cost-effective solution.
A typical mechatronics project is substantially more complicated than this and
thus less straightforward to assess, not least because it will necessarily involve
multiple Engineering disciplines and multi-skilled teams.
14.5 Teaching Mechatronics—An Example
We have comprehensively overhauled the entire Engineering undergraduate expe-

rience at the University of Greenwich. As with many newer universities, the focus
of Engineering programmes had typically been on the acquisition of technical
knowledge. Consequently, the entire pedagogical experience had been focused
on technical instruction, typically in the traditional forms of equations and laws,
delivered through lectures supported by laboratory sessions. Assessments were
largely exam-based, with traditional mathematically heavy questions where
answers were typically either correct or incorrect. Exploration of problem and
solution spaces is difficult to encourage in this context.
While the acquisition of technical knowledge is clearly a key requirement of
any undergraduate programme, the pedagogical focus on this somewhat narrow
goal tended to miss the wider objectives of preparing the students for professional
practice. In particular, important skills such as innovation, creativity and engi-
neering “instinct”, the ability to look at a design and have a realistic view of its
merits and weaknesses, were not typically taught. This apparent oversight was not
because the academic did not appreciate the value of such skills, more that the
programme structure and assessment practices did not lend themselves to support-
ing them, for the reasons discussed earlier in this chapter. Furthermore, the pro-
grammes were delivered in a heavily silo-ed approach, which made the delivery
of strongly interdisciplinary subjects such as mechatronics inherently difficult
administratively.
Given that we believe that with the rise of notions like the Internet of Things
[38], the traditional silos are increasingly archaic, we took the step of completely
re-thinking all of the programmes. A number of new degree programmes were
introduced, such as Design, Innovation and Entrepreneurship—to help encourage
the next generation of entrepreneur–inventors—and Engineering for Intelligent
Systems—which is, in effect, a degree in mechatronics.
A new common first year, focusing on the fundamental principles of
Engineering Science, was introduced for all Engineering students, whether study-
ing on traditional programmes, such as Civil or Mechanical Engineering, or the
newer programmes. The new first year consists of four double-courses:
Engineering mathematics Students explore a range of engineering problems through which

relevant Mathematical skills are taught
Practical and Students are provided with the lab sheets at the start of the year,
experimental skills complete with theoretical primers that are to be completed prior
to the lab sessions. The lab sessions then focus on “learning by
doing”, i.e. verifying the theoretical answers through replication
in the labs
Engineering professional Students are taught the wider aspects of becoming a
skills professional engineering, such as communication (including
essay writing, critiquing, how to précis and presentation skills),
risk assessment and management (including the study of
engineering failures), ethics and management, among other skills
Design and materials This consists of some traditional Materials instruction
coupled with an introduction to Engineering Design. These
complementary topics are then combined into a group design,
build and evaluate mechatronics exercise.
An example challenge is to build a remote-control boat. The students are given

a budget of £50 and are allocated a material out of which to build their hull. These
materials can vary from newspaper to plastic drinks straws or ice cream tubs. A
series of challenges for the boats to complete are set, around attributes such as
speed and manoeuvrability. For example, in any one year the challenges may
include:
• Build the fastest boat.
• Complete the obstacle course in the fastest time and with the fewest penalties.
• Be the most aesthetically pleasing.
• Be the best value-for-money.
Students then have to decide for which challenges to prioritise with their designs.
A possible grading scheme could be developed by attaching values to each of
these factors and a simple algorithm implemented to calculate a total “score” for
each group. However, once the students become aware of how the scoring algo-
rithm works, this knowledge will axiomatically influence how they approach
the design process, thus potentially stifling their creativity. For example, should
encountering an obstacle be more heavily penalised than, say, time to complete a
traverse, then the students will begin to prefer slow, but steady solutions.
232 S. Keates
DARPA addressed this issue in its self-driving car challenge [39] where the cri-
terion for success was simply that the first vehicle to cross the finish line wins.
A consequence of this approach is a wide variety of highly innovative entrants.
Similarly, the Robot Wars television programmes had an equally direct approach
to establishing the “better” design—a fight until only one robot remained and all
opposition had either been immobilised or ejected from the arena. Again, there
was a similarly wide variety of innovative designs among the entrants. We are
in the process of working with the team behind Robot Wars to establish an out-
reach programme to local schools to inspire the next generation of mechatronics
students by helping schoolchildren design and build robots to compete in Robot
Wars.
The solution that we use was inspired by the role of the jury on Robot Wars
where a panel of external experts is used to assess each finished design against
each of the stated challenges and category champions identified. Those champions
then progress through to a final round and a “champion of champions” is named as
the design that, in the opinion of the experts, best meets as many of the challenges
as possible.
14.5.1 In Summary
Engineering Design is a major element of mechatronics and can form the unifying
theme throughout such courses. However, the requirement to encourage innovation
is often in conflict with the requirements of “quality” and of the need to assign
grades to all forms of student-based activity, even when doing so encourages a
conservative approach to design. Instead, the aim should be to encourage innova-
tion, and even failure, as to reward students for the adoption of an innovative and a
novel approach.
One possible way of accomplishing this is to simplify the criteria or measures
of success as much as possible—ideally to a single such metric, e.g. the fastest or
the lightest. It is also suggested that all mechatronics programmes focus not only
on the development of working solutions, but also on how the solutions fit within
the wider environment of use, including their users.
14.6 A Final Note—Do not Forget the User
A common failing among many mechatronics projects is a focus on the technical

capability of the device or robot being constructed. This failing is not restricted
solely to students; it pervades many mechatronics industrial and research projects.
For example, the first iteration of IRVIS project [37] discussed earlier failed to
produce a usable robot. The development team had spent 3 years developing the
robot and ensuring that it functioned. The interface received scant attention until
almost the very end of the project such that when the robot was taken to the indus-
trial test site, the interface was a barely developed version of the testing interface
used to drive the motors individually. The final user acceptance test was a failure,
because although the user could move each of the motors individually, the visual
inspection task required complex simultaneous motor control, which the interface
simply did not support.
A second three-year development cycle was required to address these short-
comings. The original development team was replaced and their parting advice
to the new team was that the acceptance trials failed because the robot was
under-specified and needed a (very expensive) complete overhaul. The new
development team instead focused on developing a working interface by focus-
ing on the end tasks of the user. A more complete, task-focused interface was
developed and the user acceptance trials were completed with no significant
shortcomings being identified. No overhaul of the robot itself was required. The
deficiencies in performance suggested by the first set of user trials was a result
of the motors not being driven effectively—one at a time instead of combina-
tions together.
The experience of this project is unfortunately common among many such
mechatronics projects. In a very insightful paper, Buhler examined the success of
several of the major EU TIDE Rehabilitation Robotics projects in the 1990s [40].
His conclusion was that only one of the projects that he evaluated (the MANUS
project [41]) had achieved its original design objectives and had achieved a
respectable degree of success. All of the other projects were considered failures
and the most common reason for failure that was identified was a focus on the
technology to the exclusion of almost all other considerations.
Clearly, any mechatronics programme must bear this in mind and ensure that
students are aware not only of how to develop such systems, but also how they
interact with the wider environment, including their users. Such considerations are
routinely taken into account in other specialist domains, such as medical device
design and it is suggested that mechatronics students are made aware of such
broader approaches to Engineering Design.
IRVIS, as a mechatronics product, was very basic compared with the capabili-
ties of modern systems, such as RoboThespian, shown in Fig. 14.7. RoboThespian
has been designed explicitly to mimic human movements and appearance. Final
year students are taking up projects to explore how people may wish to interact
with the robot and it is straightforward to code and implement lifelike responses.
At the same time the success of the IBM Watson system in answering unstructured
questions in the Jeopardy!TM challenge [42] shows that “artificial intelligence” is
developing apace.
234 S. Keates
Fig. 14.7 RoboThespian
14.7 The Future
Mechatronics is moving to a future where the design of complex physical compo-

nents is becoming commoditised, i.e. it is becoming easier to find complex prod-
ucts off-the-shelf, meaning the real area for innovation is in exploring innovative
ways to use such capabilities to interact with people.
As we have seen, mechatronics is necessarily a cutting edge discipline where
technology is changing rapidly. Humanoid robots, such as RoboThespian, that
were the stuff of science fiction only a decade or so ago are now available to pur-
chase. Their cost is still prohibitively expensive for many degree programmes, but
similar technologies have shown that an order of magnitude decrease in price is
eminently achievable over a relatively short time span as the technology becomes
increasingly commoditised.
Indeed, this process of commoditisation is changing much of Engineering and
Technology education, as increasingly complex functions do not typically need
solutions to be custom built from scratch. Instead, increasingly powerful modu-
lar components can be brought together as an assembly, and with the correct set-
tings and control coding can accomplish complex tasks without students needing
to reach for the soldering iron.
While it is still very necessary that students understand what goes into each
modular component, how they are designed, and what their capabilities and lim-
its are, there is also a growing challenge in terms of the opportunities that are
now opening up. The power and potential of these systems means that engineers
and designers are now on the verge of being able to think very ambitiously about
what they would like their device or system to accomplish, almost unlimited and
unrestricted by the capabilities of the hardware. We are not quite there yet, but the
capability of the technology is now only a small step behind that of the imagina-
tion of the typical Engineering student.
The impact of the next generation of mechatronics devices is already being
felt. Take, for example, the rise of 3D printing. In the 1960s and 1970s, compa-
nies began to realise that labour costs in the developing world were very much less
than in developed countries. The notion of offshoring was born and the following
few decades saw the manufacture of low added value products in particular being
transferred from countries such as the US and UK to the Far East and elsewhere.
However, it is highly likely that the “no labour” costs of 3D printers will undercut
even those low labour costs, and also have the added advantage that the products
can be made at the point of demand and do not need shipping halfway round the
world. Once 3D printers and other similar technologies become sufficiently com-
monplace, the money to be gained in manufacturing will move from those who
can make the product most cost effectively to those who can design the most use-
ful or desirable product.
Similarly, the Internet of Things is also an increasingly important development
that has the potential to change the world in which we as much as the Internet
itself has done since the early 1990s. Again, technologies that are already available
are capable of supporting many exciting innovations. However, it is still looking a
little like a solution in search of a problem. The only innovations that have thus far
gained any notable traction in the market place are somewhat mundane, with ele-
ments of home automation, home security and heating applications initially being
the most pervasive Internet of Things solutions in the marketplace. Such applica-
tions are only scratching the surface of what the technology can support. However,
designers and developers are still struggling to find the “killer application(s)”
that will lead to sufficient homeowners investing serious money in the necessary
Internet of Things infrastructure in their house.
Changes in the general population also need to be considered. Many coun-
tries in the developed world already have populations that can be considered
aged, rather than ageing. There is a clear need for more technology to help sup-
port people in retaining their ability to maintain independent living in their own
homes [43]. Mechatronics will underpin much of the new developments in tele-
healthcare, assistive technologies and support for the activities of daily living
[38]. However, designing for older adults or those with disabilities involves par-
ticular design challenges because of the variety of user functional capabilities
[44] that may be encountered as well as different user priorities and goals [45].
Consequently, future mechatronics engineers will need to understand as much
about consumer wants, needs and aspirations as they will about, say, different
types of motors.
To reiterate what was stated in the introduction to the chapter, The goal of any
good degree programme is to not only prepare each student to secure their first job,
but also to give them the correct skills and mindsets to retain employment throughout
236 S. Keates
their entire working life, requiring educators to consider how mechatronics is likely
to change in the mid- to longer-term future, and how these changes are likely to
impact on course content, structure and delivery. This is a particular challenge in a
discipline such as mechatronics with all its diversity. The solution must be to aim for
a balance between:
• Technical knowledge—Providing sufficient content about the technology of
today.
• Underlying fundamental technical skills—Skills such as Design and
Mathematics will support graduates throughout their working life.
• Personal skills—These encompass lifelong learning, adaptability, problem-solv-
ing and open-mindedness that together make up a flexible and adaptive mindset,
open to new challenges.
References
1. Buur J (1991) A theoretical approach to mechatronics design. Institute for Engineering

Design, Technical University of Denmark
2. Bradley DA (2004) What is mechatronics and why teach it? Intl J Electr Eng Educ
41(4):275–291
3. Hanson M (1994) Teaching mechatronics at tertiary level. Mechatronics Spec Issue
Mechatronics Sweden 4(2):217–225
4. Acer M, Parkin RM (1996) Engineering education for mechatronics. IEEE Trans Ind
Electron 43(1):106–112
5. Bradley DA (2010) Mechatronics: more questions than answers. Mechatronics
20(8):827–841
6. Vantsevich VV (2010) Education in Mechatronics, Mechatronics in Action. Springer, Berlin,
pp 197–218
7. Lek CC, Yew CK (2010) Mechatronics education in Singapore, 7 p. http://mechatronics-net.
de/first-ws/ws_cheah.pdf. Accessed 20th Aug 2015
8. Grierson H, Nicol D, Littlejohn A, Wodehouse A (2004) Structuring and sharing information
resources to support concept development and design learning. In: Network learning confer-
ence 2004, Exeter UK
9. Grierson H, Wodehouse A, Breslin C, Ion W, Nicol D, Juster N (2008) An evaluation study of
a digital library of ideas: workflow model and classroom use. Intl J Digit Libr 9(1):29–39
10. Dym CL, Agogino AM, Eris O, Frey DD, Leifer LJ (2005) Engineering design thinking,
teaching, and learning. J Eng Educ 94(1):103–120
11. Bailey-McEwan M (2009) Difficulties of mechanical engineering students in developing
integrated knowledge for the cross-discipline of mechatronics: a conceptual investigation. In:
Report submitted to the faculty of humanities, University of the Witwatersrand in partial ful-
filment of the requirements for the degree of Master of Education
12. Atman, CJ, Adams RS, Cardella ME, Turns J, Mosborg S, Saleem J (2007) Engineering
design processes: a comparison of students and expert practitioners. J Eng Edu 96(4):259–379
13. Savin-Baden M (2000) Problem-based learning in higher education: untold stories, social.
Research into Higher Education & Open University Press
14. Doppelt Y (2005) Assessment of project-based learning in a mechatronics context. J Technol
Educ 16(2):7. http://scholar.lib.vt.edu/ejournals/JTE/v16n2/doppelt.html. Accessed 28 Dec
2015
15. Delyser RR, Thompson SS, Edelstein J, Lengsfeld C, Rosa AJ, Rullkoetter P, Whitman R,
Whitt M (2003) An educational brief: creating a student centered learning environment at the
university of denver. J Eng Educ 92:269–273
16. Giurgiutiu V, Lyons J, Rocheleau D, Liu Weiping (2005) Mechatronics/microcontroller edu-
cation for mechanical engineering students at the University of South Carolina. Mechatronics
15:1025–1036
17. Falchikov N, Goldfinch J (2000) Student peer assessment in higher education: a meta-analy-
sis comparing peer and teacher marks. Rev Educ Res 70(3):287–322
18. Hanrahan SJ, Isaacs G (2001) Assessing self- and peer-assessment: the students’ views. High
Educ Res Dev 20(1):53–70
19. Papinczak T, Young L, Groves M (2007) Peer assessment in problem-based learning: a quali-
tative study. Adv Health Sci Educ 12:169–186
technology. Harvard Business School Press, Boston
21. Chesbrough HW, Vanhaverbeke W, West J (eds) (2006) Open innovation—researching a new
paradigm. Oxford University Press, Oxford
22. http://secure3.verticali.net/pg-connection-portal/ctx/noauth/PortalHome.do. Accessed 20th
Aug 2015
23. www.ed.gov/open/plan/flagship-initiative-collaboration. Accessed 20th Aug 2015
24. IBM Global Innovation Outlook. www.ibm.com/ibm/gio/us/en/index.html. Accessed 20th
Aug 2015
25. Prados JW (1998) Engineering education in the United States: past, present, and future. In:
International conference on engineering education, 8 p
26. Stouffer WB, Russell JS, Oliva MG (2004) Making the strange familiar: creativity and the
future of engineering education. In: Proceedings of 2004 American Society for engineering
education annual conference, Session #1615, 13 p
27. Malmqvist J, Young PW, Hallström S, Kuttenkeuler J, Svensson T (2004) Lessons
learned from design-build-test based project courses. In: International design conference,
DESIGN2004, 8 p
28. Berggren K-F, Brodeur B, Crawley EF, Ingemarsson I, Litant WTG, Malmqvist J, Östlund
S (2003) An international initiative for reforming engineering education. World Trans Eng
Technol Educ 2(1):49–52
29. Crawley EF, Malmqvist J, Östlund S (2007) Rethinking engineering education: the CDIO
approach. Springer, Berlin
30. Schaefer D, Panchal JH, Choi S-K, Mistree F (2008) Strategic design of engineering educa-
tion for the flat world. Intl J Eng Educ 24(2):274–282
31. Colgate JE, Bruce AM, Ankenman B (2004) Implementing design throughout the curriculum
at northwestern. Design Eng Educ, 12 p
32. Platanitis G, Pop-Iliev R (2010) Establishing fair objectives and grading criteria for under-
graduate design engineering project work: an ongoing experiment. Intl J Res Rev Appl Sci
3(5):271–288
33. Wodehouse A, Bradley DA (2003) Computer tools in product development. In: Proceedings
of 14th international conference on engineering design, ICED03:DS31_1096FPC
34. Macintyre FR, Thomson A, Wodehouse A (2011) Identification, translation and realisation of
requirements for a knowledge management system in an engineering design consultancy. In:
Proceedings of 18th international conference on engineering design, ICED 2011
35. www.arduino.cc/. Accessed 20th Aug 2015
36. http://mindstorms.lego.com/. Accessed 20th Aug 2015
37. Keates S, Clarkson PJ, Robinson P (1999) Designing a usable interface for an interactive robot.
In: Proceedings of 6th international conference rehabilitation robotics (ICORR99), pp 156–162.
www.rehabrobotics.org/icorr1999/papers/papers/keates.pdf. Accessed 20th Aug 2015
238 S. Keates
38. Keates S, Bradley D, Sapeluk A (2013) The future of universal access? Merging computing,
design and engineering. Universal Access in Human-Computer Interaction. Applications and
Services for Quality of Life, Springer, Berlin, pp 54–63
39. Seetharaman G, Lakhotia A, Blasch EP (2006) Unmanned vehicles come of age: the DARPA
grand challenge. Computer 39(12):26–29
40. Buhler C (1998) Robotics for rehabilitation—a European(?) perspective. Robotica
16(5):487–490
41. Driessen BJ, Evers HG, van Woerden JA (2001) MANUS—a wheelchair-mounted rehabilita-
tion robot. Proceedings of IMechE H 215(3):285–290
42. Keates S, Varker P, Spowart F (2011) Human-machine design considerations in advanced
machine-learning systems. IEEE/IBM J Res Dev 55(5):1–4:10
43. Keates S, Clarkson PJ (2003) Countering design exclusion: an introduction to inclusive
design. Springer, Berlin
44. Clarkson PJ, Keates S, Dong H (2003) Quantifying design exclusion, in Inclusive design:
design for the whole population. In: Clarkson J, Coleman R, Keates S, Lebbon C (eds),
45. Keates S (2006) Designing for accessibility: a business guide to countering design exclusion,
Human factors and ergonomics series. L Erlbaum Associates, Abingdon
Chapter 15
Mechatronics Education:
Meeting Future Need
David Russell
15.1 Introduction
Of all the topics feverishly discussed in staff meetings, conferences and blogs,
future educational effectiveness and relevance always seems to crop up.
Professors, remembering their own education and struggles to attain their current
status in academe bemoan the current lack of mathematics, the lack of student tal-
ent, and the drift away from hard design to a plug-and-play mentality. There are
evident disjoints between Software-as-a-Sservice (SaaS), cloud computing, and
Platform-as-a-Service (PaaS) and mechatronic systems.
While abstraction from the internals of the computational units that constitute
the Internet of Things (IoT) might speed up product launches, the mechatronic
engineer who is operating in the actual application domain is left pondering the
integrity of the software and its source, its ruggedness over time in the real world
setting, how to manage component upgrades, and the recovery of the system after
a failure. This chapter includes true vignettes, disasters, challenges and discus-
sion topics from the author’s experience, selected to highlight what a mechatron-
ics engineer must know and to illustrate the necessity for innovation and technical
dexterity. Each subsection of this chapter is chosen to highlight technical and non-
technical topics that should be integral to mechatronics education long into the
future.
The author was an invited panel member in the vigorous discussion at the
Mechatronics 2014 conference held in Karlstad, Sweden in June 2014. He has
worked in the manufacturing systems integration industry and academia for almost
fifty years. The views expressed in this chapter are his alone, and designed to pro-
voke discussion and hopefully bring about real advances in mechatronics edu-
cation among teaching staff and administrators in the institutions of its readers.
As educational delivery mechanisms migrate from the traditional lecture-recitation
D. Russell (*)
Penn State Great Valley, Malvern, USA

DOI 10.1007/978-3-319-32156-1_15
240 D. Russell
classroom in favour of more outcome-based syllabi and technology-enhanced

learning, it is hoped that the reader will be able to decide upon the best course of
action to take for mechatronics and similar discipline courses of study.
15.2 The Educational Experience and Employment
Taking a rapid global scan of the educational process, it is obvious that there is
no real body of knowledge of mechatronics as opposed to say the medical profes-
sion. It is not within the purview of this chapter to compare countries with coun-
tries, universities with universities or even precollege common core education. The
objective is to highlight how differently somewhat similar materials can be taught
to the students who constitute the future engineering cohort yielding to the inevita-
bility of on-line delivery.
While on-line instruction, at the time of writing, may be in an ascendancy, the
Center for Teaching and Learning at the University of North Carolina Charlotte
(UNCC) [1] lists 150 different teaching methods, admittedly not all of which
apply to mechatronics. These range from the well-known “lecture by the teacher”
which appears as #1, to “small group brainstorming” listed as #150. Buried as
#106 is “the use of technology and instructional resources.” At the risk of being
facetious, the chapter author’s favourite is #127 “visit an ethnic restaurant.” But,
what is best for the student?
There are many instructional methods. Table 15.1 is based on a College of
Southern Nevada (CSN) website [2] and summarizes some of the instructional
methods that can be affiliated with the various teaching styles.
Academic readers will readily associate how classes at their institutions are
conducted in the main. Following the full CSN website, the interested reader may
find how these methods translate to an on-line environment interesting.
15.2.1 The Institution
In the US, there are over one thousand colleges and universities that boast hav-
ing an engineering school. This number is increased significantly if the number of
engineering departments in Europe, China and India are added. Most schools are
regulated by governing bodies (e.g. ABET in the US) as far as the curriculum is
concerned, but there is no common core curriculum for the nation.
This means that what is taught at Institution A may be covered superficially
or not at all in Institution B. Overseas, the problem is worse. Some engineering
schools in certain countries do not pass muster outside of the country itself. By
awarding engineering degrees such institutions promise good jobs and better lives
for their graduates only to not even be considered for a good job inside or outside
the country. This is not good for the student.
15 Mechatronics Education: Meeting Future Need 241
Table 15.1 Instructional methods and teaching styles

Method Comments
Lecture A flexible method which can be applied to almost any
content. Although lectures can be very engaging, they put
students in a passive role. Experienced staff members can
interweave their real-world experiences into course materi-
als to show the relevance of the class
Teaching Style—Formal Authority
Lecture-discussions Combines the lecture with short question periods or a
series of short question periods for students
Teaching Style—Formal Authority
Demonstrations Involves students learning a process or procedure based on
instructor performance. The students may be involved in
the demonstration and practice
Teaching Style—Demonstrator
Simulations Simulations put learners into seemingly real situations
where they can make decisions and experience the out-
comes of their decisions without the risk
Teaching Style—Facilitator/Delegator
Collaborative learning Students process information and derive knowledge
through discussing course-related issues and topics with
each other
Teaching Style—Facilitator
Cooperative learning Small groups of students work together to solve a problem
or complete a task
Case studies This involves individuals or groups of students working
together to analyze a case, which is customarily a real-life
situation which has been written up to highlight problems
and solutions
Role play Students work to solve problems through adopting the
different roles associated with it. Role play involves
identifying, acting out, and discussing problems. With care
this can be highly effective especially in the non-technical
aspects of systems engineering such as human resource
management
Problem based and inquiry Instructors give students a problem which the student must
learning solve by gathering data, organizing data, and attempting
an explanation. Students should also analyze strategies
that they used to solve the problem
Teaching Style—Formal Delegator
To rectify this problem, many well regarded colleges and universities are
populating on-line and residential post-graduate courses. Mechatronics, robotics
and other disciplines are popular topics in what are intended to be educational
objects.
242 D. Russell
15.2.2 The College Faculty and Staff
University teaching staff, instructors, and professors ideally are mature and have
some actual industrial experience. With no real pedagogic training, they teach as
they were taught, with much theory and arguably little relevance to their students
interests or final occupations. Most teaching staff have had little or no formal
training in teaching, classroom management, or legal and ethical matters.
US News and World Report ranks the top schools annually but this ranking gen-
erally reflects research expenditures, the number of doctoral degrees awarded if
appropriate, a tally of staff who hold a terminal degree and Fellow status within
their institution. The rating may include graduation and retention rates. Teaching
may be prescribed for each staff member, but it is certainly held in lower regard
than funded research in contract renewal matters.
Efforts such as the UK Teaching Quality Assessment (TQA) were designed to
highlight and reward good teaching practice at schools and colleges in much the
same way the Research Assessment Exercise (RAE) handles research. It is the
responsibility of university staff to perform both research and teaching well to
promote high marks in both the TQA and RAE reviews. In the US, engineering
departments are subject to a periodic nationwide ABET accreditation process but
only at the baccalaureate level. But, what is best for the student?
15.2.3 The College Student
In the US for example, many engineering students spend just over two years in
fairly focussed programs (e.g. electrical engineering) and may select their major
while in their first or second year. Concurrent with these studies, students will be
exposed to ethics, legal issues and presentation. In Europe, students may enter
programs already knowing their chosen field and experience four years of topical
study. Some schools inject a term of work experience before their final year while
others engage final year student projects.
It almost goes without saying that successful students will have good study
skills and an excitement about engineering while lackadaisical students tend
to do poorly and often transfer into other (self-perceived as easier) programs or
institutions. It is a well-known construct that how a student learns about Science,
Technology, Engineering and Maths (STEM) before college is a major indicator
as to what fields of study the college-bound student selects; this varies globally
as will be shown at the end of the chapter. Despite scholarships and financial aid,
location, need and social status do figure as to which institutions are feasible to an
applicant.
Engineering schools worldwide are somewhat selective and require four or
even five years of study for a baccalaureate degree. Mechatronics is certainly
taught at the baccalaureate, masters, and doctorate levels but usually championed
by enthusiastic staff. Are students attracted to post-graduate degrees to help staff

with research and teaching rather than industrial employ? Is this best for the
student?
15.2.4 The Mechatronics Employer
Imagine now that the student has successfully managed to gain employment in a
technical company that for the sake of this chapter produces or uses mechatronic
systems. Such employers have a perceived need for expertise to further their prod-
uct or service and have high expectations for the incoming graduate or technician.
In the legal and medical professions, novitiates must complete residencies to
become certified before being allowed to practice, whereas in engineering char-
tered membership in an institution is considered largely optional, expensive, and
irrelevant. It is common practice for new employees shadow experienced engi-
neers until they can be assigned to projects experts in their own right. From this,
the reader can deduce why projects fail, how cost overruns happen and products
never quite work as anticipated by the client. What is best for the company?
15.3 Mechatronics: A Selection of Real World Vignettes
The following contains three factual real world vignettes from the chapter author’s
experience designed to reflect necessary topics in mechatronic education. The cor-
poration or company names are omitted for confidentiality reasons but hopefully
the reader will find the examples useful. Each subsection will briefly describe a
real system and how it was designed, how a problem presented itself, the reso-
lution of the problem, and most importantly, what educational skill enabled the
mechatronics engineer to address the problem. The first case is given in much
more detail than the other two to better illustrate the point.
15.3.1 An Injection Moulding Monitoring System
Overview
An injection moulding corporation is contracted with a systems engineering com-
pany to design and implement a production monitoring system for its main loca-
tion that operates up to 40 high-tech moulding machines. About 35 machines run
regularly on any one day producing several tens of millions small plastic parts
daily. The components are packed in boxes by weight and passed on to qual-
ity control and inventory. Figure 15.1 shows a typical injection moulding (IM)
factory.
244 D. Russell
Fig. 15.1 A typical US injection moulding operation (Courtesy of the Rodon Group, Hatfield PA)
The factory manufactures a variety of items on a job-by-job basis. A job change

on any machine requires much effort in purging the previous, coloured raw mate-
rials and the necessary mould mounted, and new liquid plastic bled through the
system for the next job. The mechanic may cycle the machine many times until the
new part is perfect, but these test operations should not ever appear in the produc-
tion count.
Summary of Requirements
Without going into further detail, the requirements of the system included the
measurement of each cycle of each machine on a 24 × 7 basis, a comparison of
actual performance with the factory work order, the provision of display screens
throughout the factory and the periodic download of inventory data to a mainframe
computer. From a data integrity standpoint, this is actually very difficult to do
because not all machine cycles produce product, e.g. a technician loading a new
job or clearing a mould jam.
System Design
After meeting with the industrial client several times, Fig. 15.2 emerged as the
preliminary system design. The major components are fairly standard in most
industrial automation setups. The programmable logic controllers (PLC) are indus-
trial process control agents which are resistant to power outages and are available
with local storage, communication capabilities and multiple input and output data
ports.
Having designed the system the following hitherto unforeseen questions were
posed after a more detailed system site inspection:
Fig. 15.2 Preliminary system design
1. How to connect machine information over long distances? The factory is over a
mile long.
2. How to connect all of the system devices over such long distances? Electrical
signals were all low quality with much apparent noise being generated ran-
domly from the injection presses.
3. How much information is it useful to display?
4. How can operators and mechanics provide specific data for display?
Once these matters were resolved, which in fact did involve some redesign of the
system and the purchase of additional software and hardware, the system was
coded and installed.
Problem Areas
In the day to day operation of the system, the following unexpected situations
arose:
1. What appeared to be random data freezing anytime during operation.
2. Data loss after a blackout or brownout of primary factory power.
3. Handling machine maintenance and repair status cycles.
4. Shift reports show incorrect times.
These problems seemed to indicate fatal flaws in the system, yet were solvable
using mechatronic principles. The chapter author’s solutions are summarized in
Sect. 15.5.1.
15.3.2 Executing Mainframe Code on a Minicomputer
Overview
A company was using a mainframe computer for advanced CADCAM and graph-
ics. Each design station cost over $50,000 and the mainframe lease and operating
246 D. Russell
system was over $100,000 per month. The consultant found a company that had
found a way to run instructions from the mainframe on a $20 K minicomputer by
making some minor adjustments to the motherboard of the minicomputer.
Overview of Invention
Figure 15.3 illustrates how the mainframe instructions were accessed and exe-
cuted by the minicomputer by modification of the minicomputer motherboard
with proprietary firmware. The schematic blocks shown dashed were the only
firmware modifications needed. The minicomputer word size must be compara-
ble with the mainframe instruction chip set (32 bit) which was purchased from the
manufacturer.
Problem Area
The system functioned very well and the CADCAM application was successful
and an inexpensive alternate to the traditional graphics workstation. One day, after
a minicomputer operating system upgrade, the system completely failed to oper-
ate. Mainframe computer instructions embedded in the CADCAM sequences sud-
denly caused the minicomputer to return an illegal instruction trap and a complete
CADCAM failure.
This problem indicated a fatal flaw in the system that eventually proved unsolv-
able causing the project to be discontinued. The chapter author’s explanation is
summarized in Sect. 15.5.2.
Fig. 15.3 Modified minicomputer motherboard schematic

15.3.3 A Mechanically Unstable System
Overview
Many researchers have studied various methods of inducing control into an
inverted pendulum rig. This system lends itself to adaptive, intelligent, evolution-
ary and learning control. Figure 15.4 is a photograph of one such rig with which
the author worked [3]. Essentially, the cart was driven in bang-bang LEFT/RIGHT
mode on computer command. The experiment was bounded on a two meter track
with crash sensors at each end. The pole on the cart was freely hinged but lim-
ited to about ±10°. If the system went out of range, the motion on the cart was
stopped. The problem was to balance the pole by moving the cart left or right and
should not be confused with the swing up pole balancing act.
Problem Areas
The two major problems were ensuring that the system engaged its learning algo-
rithm from an initial random but legal state so that the controller could recognize it
and launch out on a control excursion, and handling slippage in the driven wheels
when the cart direction was reversed. The chapter author’s explanation of a solu-
tion to the first problem is summarized in Sect. 15.5.3.
15.3.4 Summary of Cases
For each of the above three cases, how these problematic situations were
addressed appears below in Sect. 15.5 to encourage readers to discuss their own
ideas with those of their students before reading that section. After reading the
author’s comments, readers should discuss then what educational modules at their
institution or company would have enabled the novitiate engineer to address those
problems?
Fig. 15.4 A trolley and pole

experimental rig
248 D. Russell
Perhaps the missing educational experiences in our colleges and universities are
in-depth coverage of systems engineering and system integration.
15.4 Systems Engineering and Systems Integration
In the cases given above in Sect. 15.3, it should be apparent that the designs of the
system components, the integrated system, and even the placement of the system
within its global domain (a.k.a. in a system of systems) rely heavily on the under-
standing of systems engineering and systems integration.
15.4.1 Systems Engineering
Perhaps the clearest definition of systems engineering is found on website [4] of

the International Council on Systems Engineering (INCOSE) from where the fol-
lowing quotes are taken:
(INCOSE) … represents systems engineering professionals from industry, government,
and academia worldwide. It strongly believes that the fundamental principles of systems
engineering have an important role in the education of all engineers, regardless of their
specialty, as well as professionals who work with systems engineers but do not have an
engineering background.
The same website explains the nature of the discipline and its truly outcome-
based focus.
Systems engineering is an interdisciplinary approach and means to enable the realization
of successful systems. It focuses on defining customer needs and required functionality
early in the development cycle, documenting requirements, then proceeding with design
synthesis and system validation while considering the complete problem.
Specifically, systems engineering is an integrative paradigm that for many years

was never taught in engineering colleges, assuming that graduates of their pro-
grams will pick this up later in their careers.
Systems engineering integrates all the disciplines and specialty groups into a team effort
forming a structured development process that proceeds from concept to production to
operation. Systems engineering considers both the business and the technical needs of all
customers with the goal of providing a quality product that meets the user needs.
15.4.2 System Integration
System integration is a well-known subject in computer science and IT and has

come to mean the assembly of software systems using plug-and-play paradigms
from software located as COTS (Code off the Shelf), SaaS (Software as a Service)
and of late Cloud Services. In the software realm, the major integration issues in
an open architecture environment are system and application configuration. In this
activity, the integrator has to skilfully slot the application into sets of code that
may have been written externally in another language. Enterprise software systems
such as SAP® require the use of many configuration forms and data manoeuvres
before a manufacturing company can benefit from its complexity and information
power. Most problems arise from hardware failures, internet issues and misfits in
terminology and usage.
In mechatronic engineering systems integration problems arise from combi-
nations of mechanical, electrical, computer and systems disciplines. Solving in
one area may cause sudden failure on another front. The second Sect. (15.3.2) is
an illustration of how a project failed through no fault of its own as explained in
Sect. 15.5.2. It was actually the reluctance of the minicomputer vendor to make a
simple revision to their operation system that caused the failure.
Computer engineering and computer science programs usually include some
information integration, database and internet-enabled modules. Formal engi-
neering programs by and large contain very little curricular coverage of system
integration. Warminki and Ikonomov [5] opine that: “… the basic engineering
curriculum fails to teach valuable skills in the areas of:
• Knowledge management/documentation/recall and reuse.
• Working in cross functional distributed teams.
• Critical thought in the framework of product design.
• Design methodology including: translation of vague requirements to engineer-
ing specifications, failure mode identification and effect analysis, total param-
eter and tolerance product design, manufacturing execution, function as a
member of a team to undertake the analysis and integration of automated manu-
facturing processes.”
This issue at their institution is being addressed by a detailed hands-on project in
which students are posed with real problems to solve. In a group situation, stu-
dents can engage in problem solving activities such as Scrum [6] and other similar
team oriented project work.
15.4.3 Hands-on Versus Knowledge-Based Instruction
This project-based approach introduces the controversy of the educational value of

hands-on “tinkering” by students versus a traditional solid educational classroom
instruction. The popular vogue of “learning by doing” may work well in simple
classroom situations, but would it work in the cases given above in Sect. 15.3?
Can an impatient, paying, client be expected to wait for expertise to be learned?
Section 15.2.4 is understated.
250 D. Russell
Formal engineering programs especially those under accreditation control are

loath to forfeit more classical topics in favour of mechatronics or systems engi-
neering. Many schools have introduced one or two year taught master’s programs
in mechatronics. These are more popular in the US than in Europe. In all, it is the
experience, enthusiasm and focus of university staff that are charged with the trust
of producing ethical, worldly wise competent engineers of all disciplines. Much
project work is undertaken on an individual level with little interface with other
students, whereas in industry the ability to work in a team is a much sort after
skill.
15.5 Solutions and Educational Sources to Case Issues
The following are outlines of how the problematic areas of each case were
resolved, but readers may want to discuss other solutions with their colleagues and
classes. Much more detail is given to the first case to illustrate the complexity of
mechatronic systems and because it was housed in a real-world industrial environ-
ment. The second focussed on the need for a fairly deep knowledge of operating
systems and firmware, and the third on mechanical design and the use of timed
software.
15.5.1 An Injection Moulding Monitoring System

(Case 15.3.1)
The solutions to the problems introduced in Sect. 15.3.1 are summarized below
but it should be clearly understood that this is not an exhaustive list.
Problem (a) and (b)
These questions focussed on the long distances connecting devices and the low
quality and high noise electrical signals.
Solution—The use of shorthaul modems and a check on all wire shielding in
the factory roof helped with this problem. A better, if more expensive, solution
would have been to rewire using fibre optic cables.
Educational Objects—The engineer needed to be conversant with modems,
communications and fibre wire connections over long distances.
Problems (c) and (d)
These introduced the issue of good data collection, displays and factory floor
inputs.
Solution—It is essential that a focus group that includes the industrial client
and factory floor personnel decide what data is to displayed on the shop floor.
It became apparent in the system in question that shop floor data needed to be
collected from the operators. This data then identified the need and nature of a
machine breakdown, etc. It was necessary to instal microterminals and integrate
this data into the database using data fusion techniques.
Educational Objects—The system designers needed a deep understanding of
database design and fusion, and human computer interaction.
Problems (e), (f), (g) and (h)
These all occur during the operational phase of the system from time to time. In
the original system, the data collection and all database operations would freeze
mimicking the effects of a power outage.
Solution—The design and implementation of the factory software required
a level of system intelligence so that temporary problems and failures could be
detected and “self healed” to avoid loss of data. The actual system included pro-
grammable logic controllers (PLC) in which front end intelligence was embedded
to temporarily store data during a system pause or stoppage.
Educational Objects—The mechatronics engineer needed to understand file
locking and system programming to free locked files and folders. Real time oper-
ating system design knowledge is essential as was a familiarity with available
industrial components.
15.5.2 Executing Mainframe Code on a Minicomputer

System Failure
How the system works

Figure 15.2 depicts how the proprietary firmware purchased for modification of a
minicomputer motherboard utilizes an unused bit 17 in the minicomputer’s 32 bit
processor status word (PSW). The operating system kernel allowed system users
to access all PSW bits in high priority tasks. Included in the PSW is a bit 3 that
traps an instruction error. It was this bit (bit 3) that is set when the minicomputer
attempted to execute a mainframe instruction. If the executive program detects
such an event, it sets what was the last unused bit (bit 17) that was designed to
direct execution to the additional hardware for execution.
Reason for failure
The minicomputer vendor issued an update to the operating system that innocently
used that bit (17) for a new elaborate print function. The operating system soft-
ware team had spent many hours developing this new function that would benefit
all of its other customers. The CAD/CAM project was cancelled.
Educational Objects—For the mechatronic engineer to detect, this would
require a fairly high level of computer architecture, systems programming,
and firmware. As an aside, advanced negotiating skills might have saved the
project!
252 D. Russell
15.5.3 A Mechanically Unstable System
Randomized but Legal Initial State System

Many pole and cart systems begin with the pole being held vertically near the cen-
tre of the track. Upon release, the system is engaged and the process proceeds but
always from nearly the same initial state variable values. This is a real flaw in the
system. In the case in question, in order for the trolley and pole logic to engage
its learning paradigm from a random but recognizable initial state, it was neces-
sary to construct a startup subsystem that drove the cart in one direction for some
random time and then reversed the cart direction for a shorter random time and
then reversed it again. This would jerk the pole from its initial steady state rest-
ing position into a dynamic state but will not allow it to gain enough momentum
to fail. During the startup process the control system monitored the state variables
When the starting system entered a state in which the system’s bang-bang control-
ler value coincided with the startup value, the startup logic was disconnected in
favour of the system.
There many other such examples where readers may choose to insert their own
examples from their own experience using this approach.
15.6 Conclusion: A Global Problem with Local Solutions
Addressing future educational methods “The answer is not to be found by looking

in the rear view mirror” So states Marshall McLuhan quoted in a recent Educause
article by Brown [7]. Brown discusses concepts such as Adaptive Learning
Technologies, Learning Spaces, Learning Analytics, and Next Generation
Learning Management Systems and focuses on how students must navigate a way
through a pathway or swirl of instructional experiences.
This certainly has elements of truth but might be an oversimplification.
Engaged faculty who are able to bring their research or other technical interests
into the classroom can not only hold the attention of their class, but also create
a learning environment that causes students to be life-long learners, ethical, and
innovative. Looking again at Sect. 15.5 where plausible (and actual) solutions are
listed, readers should consider where these skills are being taught at their own
institutions.
This matter is not limited to North America or Europe, but is a global malaise
in what are often classified as good institutions in China, India, Singapore,
Australia and many other countries.
A ready solution might be a better understanding and use of continuous pro-
fessional education (CPE) modules such as offered by universities and the pro-
fessional institutions such as the IMechE, IET, IEEE, ASME and the like. Such
programs can help retrain more senior engineers as well as fill in the gaps in new
hires. For a deeper coverage of mechatronics many institutions are offering taught
masters programs which can be face to face or on-line. In these programs, stu-
dents are already degreed engineers and therefore can focus on mechatronic issues
such as described in this chapter without much mathematical or basic engineering
review.
The intent of this chapter has been to introduce some concepts of how
mechatronic systems posit a variety of problems for which students, even at the
doctoral level, may have had no in-depth instruction and who do not yet possess
the savvy of an experienced engineer. Reference to statistical data has been largely
avoided as numbers change so rapidly from year to year and are provided by unre-
liable sources.
References
1. Teaching Methods. www.teaching.uncc.edu/learning-resources/articles-books/best-practice/

instructional-methods/150-teaching-methods. Accessed 1 July 2015
2. Different Instructional Methods. www.csn.edu/pages/2359.asp. Accessed 13 July 2015
3. Russell DW, Rees SJ, Boyes JA (1977) Self organizing automata. In: Proceedings of 1977 con-
ference information and system sciences. Johns Hopkins University, Baltimore, April 1977
4. www.incose.org/AboutSE/SEEducation. Accessed 20 Oct 2015
5. Warminski F, Ikonomov P (2007) Teaching manufacturing systems integration in educational
institutions through hands-on experience. In: Proceedings of 2007 spring conference american
society for engineering education north central section, West Virginia Institute of Technology
(WVUTech), March, pp 30–31
6. Scrum Methodology & Agile Scrum Methodologies. www.scrummethodology.com/. Accessed
5 July 2015
7. Brown M (2015) Trajectories for digital technology in higher education. Educause July/
August:16–27
Chapter 16
Conclusions
Peter Hehenberger and David Bradley
The authors hope that readers of the book have enjoyed the fundamental research
topics and future visions of the contributors to the text. This chapter reflects the
interaction and integration of these major topics and tries to summarize the key
statements.
It should be pointed out that the continuous (r)evolution of the technical
(mechatronic) systems with the deeper integration of multiple disciplines (e.g. IT
functionalities and components) and the detailed consideration between the prod-
ucts and their related production processes are parts of the major trends in product
design. Furthermore, the involvement of several partners (all over the world) and
the challenges of new business process play an important role.
16.1 Global Trends and the Impact on Mechatronics
An overview on global (mega) trends has been provided by several institutions in

recent years (e.g. see [1]). They differ from each other in detail, but the major top-
ics are involved in all published studies as follows:
• Demographic change (and ageing society, healthcare systems)
• Mobility
• Globalization (and changes in the work world, economy, finance)
• Urbanization (and individualization)
P. Hehenberger (*)
Johannes Kepler University, Linz, Austria
D. Bradley

DOI 10.1007/978-3-319-32156-1_16
256 P. Hehenberger and D. Bradley
• Climate change and environmental change (and Energy and resources,

sustainability)
• Knowledge-based society (and ubiquitous intelligence, digital culture)
For manufacturing production in 2030 the following four major topics are dis-
cussed by Westkämper [2]:
• Innovative products and processes
• Knowledge-based manufacturing engineering
• New business models in the life cycle of products
• Infrastructure and education
The result of all these trends is also that the technology has to push forward. So
there are high product development potentials in mechatronics due to the combina-
tion of multiple disciplines, as discussed in the chapters of this book.
One major part of economic success is the development of innovative products
and processes. The term innovation comprises invention, introduction and sales
of a new product, service or procedure [3]. This does not only include the whole
marketing process, but also the social and economic impact. Irrespective of the
quality of the invention many, factors influence the growth of an invention into
an innovation. The main factors discussed in this context can be split into three
groups of technological, economical and social influences (see global trends [1,
2]). The field of mechatronics is known to be the source of numerous innova-
tions. However, most new developments are identified as incremental innovations.
Conceptual design has been identified as the most critical phase in product design
in context with radical innovations as the main portion of success will be estab-
lished there.
Decisions made in this early stage have a superior influence on the future
development of the product. Therefore, the definition of the requirements on
the system level for the overall product is crucial. The requirements which are
defined on the system level should reflect the customer’s wishes. To ensure that
the system under consideration meets the requirements, it is necessary to trans-
late them into properties of the solution. The development of systems merges
solutions from disparate engineering disciplines, like mechanical engineering,
electrical engineering, control engineering, etc. So it is very important to dis-
tinguish between properties which can only be assured on the system level and
those which can be assured by a single engineering discipline. Hence, it is impor-
tant to assign the different properties to the relevant level [4]. For achieving this
task models on the different hierarchical levels are necessary (discipline-specific
model and system models). From this point of view the modelling, simulation,
evaluation and optimization of the considered specific aspects are key points for
future mechatronic systems design, as is also mentioned in several of the previous
chapters.
16 Conclusions 257
16.2 Mechatronic Futures Map
It is understandable that it is not possible to discuss all aspects of Mechatronic

Futures in “one” single book. The goal in this book is how to group the challenges
into main subjects and present specific aspects from different viewpoints. The
common viewpoints and perspectives are identified below, while Fig. 16.1 shows
the map of the following topics:
• Issues and Challenges
The main driver for future evolution of mechatronic systems is the reduction
of development costs and time as well as the improvement of the designed
products using new technologies. This deals mainly with the virtualization of
the product to improve its architecture design, its verification and validation,
its production or operation. Indeed, virtualization enables more flexibility in
the different stages of the development at lower cost. The interaction between
the designed product and the production systems plays an important role in the
direction of Industry 4.0 (or Smart Manufacturing, Cyber Physical Production
Systems, etc).
• System Design, Modelling and Simulation
Mechatronic products gain a more complex structure and will have more comput-
ing power and network connectivity. This leads to the extended design challenges
in understanding the difficulties of complex systems where simulation will be a
Fig. 16.1 Mechatronic futures map

258 P. Hehenberger and D. Bradley
key technology for mastering these. The future trends, methods and models for
the design processes of mechatronic systems have to be considered as unques-
tionable enablers for transformation of complex systems into cyber physical sys-
tems or the global integration of the internet of things. These design processes for
mechatronic engineering have to support the development of the new services or
the implementation of an industrial internet for the factories of the future.
• Manufacturing Technology
Future technologies (e.g. Additive Manufacturing or AM) for physically creat-
ing mechatronic devices and systems will enable new possibilities in the design
process. There will be a shift from “design for assembly” approaches to “direct
manufacturing” approaches. So we could remove the need for post-fabrication
assembly and use of fastenings, yielding rapid production of robust devices.
Nowadays typical examples are 3D printed sensors, 3D printed electronics and
integrating multiple materials, which is the basis for the production of “Fully
Integrated Mechatronic Devices.”
• Internet of Things and Cyber Physical Systems
The current trend in mechatronics involves the deeper integration of computa-
tion and physical processes in networked mechatronic systems, cyber physical
systems (CPS) or Internet of Things (IoT). Therefore communication, inte-
gration and data analysis are considered essential since the scope for IoT will
depend upon the consolidation of diverse systems and standards, with “lower
level” (local) systems talking to each other and to “upper level” (global) sys-
tems. Typical applications are home automation, production, transport, energy,
health care and agriculture. The lauded potential social and economic benefits
are plausible but not guaranteed yet.
• Communication and Information Technologies
The key issues here are associated with the need to facilitate the forma-
tion of multidisciplinary partnerships. Without such partnerships, opportuni-
ties to innovate in both product design and service delivery may well be lost.
Consequently, organizations that establish robust forms of partnership work-
ing are more likely to secure a competitive advantage. Associated with this is a
requirement to establish new methods of securing and managing user consent,
while ensuring no economic or functional disadvantage if a user opts out of data
sharing.
This leads in turn to questions as to how to educate and inform technologists,
along with legal/jurisprudence practitioners, so that legal and societal demands
are addressed in the development, implementation and application of new
technologies.
• Mechatronics Education
Mechatronics is moving to a future where the design of complex physical com-
ponents is becoming commoditized. The particular challenge is that of sub-
ject diversity and mechatronics education must therefore aim for a balance
between “Technical knowledge”, “Underlying fundamental technical skills” and
“Personal skills” and any educational programme should be oriented to support
these areas. Typical topics, which have to be covered by mechatronics courses
16 Conclusions 259
are aligned along the product life cycle, including innovation, creativity, sys-
tems thinking, engineering and integration used a combination of project- and
problem-based learning methods. Mechatronics education is then the base for
applying newly available technologies.
References
1. Megatrends www.z-punkt.de/en/. Accessed 1 Dec 2015

2. Westkämper E (2014) Towards the re-industrialization of Europe, a concept for manufacturing
for 2030. Springer, Berlin
3. Norman DA, Verganti R (2014) Incremental and radical innovation: design research versus
technology and meaning change. Des Issues 30(1):78–96
4. Hehenberger P (2014) Perspectives on hierarchical modeling in mechatronic design. Adv Eng
Informatics 28:188–197

Tendencias Tecnologicas - Mecatronica

Uploaded by

Copyright:

Available Formats

Tendencias Tecnologicas - Mecatronica

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tendencias Tecnologicas - Mecatronica

Uploaded by

Copyright:

Available Formats

Peter Hehenberger · David Bradley

ISBN 978-3-319-32154-7 ISBN 978-3-319-32156-1 (eBook)

Library of Congress Control Number: 2016940131

© Springer International Publishing Switzerland 2016

Printed on acid-free paper

This Springer imprint is published by Springer Nature

It is an honour and a pleasure to write a foreword for the Mechatronic Futures

2000: Georgia Institute of Technology, Atlanta, USA

• Sports Technology: The Role Of Engineering In Advancing Sport, Dr. Andy

Prof. Philip Moore

12 Home Technologies, Smart Systems and eHealth. . . . . . . . . . . . . . . . . 179

About the Editors

Peter Hehenberger is an assistant professor and deputy head at the Institute of

Nicolas Albarello is a researcher in the Systems Engineering Department of Airbus

Alexandre Arnold is a research engineer at Airbus Group Innovations in the

systems makes it extremely challenging to decipher the mechanisms involved—a

In 1999, he obtained a Ph.D. in Computer-Integrated Manufacturing from the

Simeon Keates is professor and deputy pro-vice-chancellor of the Faculty of En-

school and University of Toulon, respectively, and a Ph.D. in Mechanical Engineer-

Stanislao Patalano is an associate professor in Design and Methods of Engineering

spaces to find survivors of natural disasters; gathering information on terrorist and

Marcelo Saval-Calvo received the degree in Computer Engineering in 2010 and

innovation projects, concept scouting as well as total turn-key projects. The

Xiu-Tian Yan is professor in Mechatronic Systems Technology, the director of the

David Bradley and Peter Hehenberger

© Springer International Publishing Switzerland 2016 1

Fig. 1.1 Cloud-based structures for the Internet of Things

Fig. 1.2 Cyber-physical systems

Table 1.1 Cloud functions

Fig. 1.4 Mobile phone use (after [13])

• Complexity and Ethics

Conventional approaches to engineering design typically follow a path such as

Fig. 1.6 Simplified V-Model

technologies are being explored, perhaps without necessarily a full understanding

1.2.2 Privacy and Security

Table 1.2 Perceived threats Threat Probability (%)

1.2.3 Complexity and Ethics

Fig. 1.7 An ageing population in Europe (after [32])

• Ensuring fair access to technologies—Controlling access to technology can act

1Similar data can be found for other global regions.

devices provides opportunities for developments in areas such as eHealth and

There is a recognised need to move towards more sustainable forms of society-cen-

Fig. 1.8 Sustainability domains

Table 1.3 City problems (after [39])

in achieving sustainability is the effective management and use of all resources;

Mechatronics education has always faced the challenge of balancing appropriate

Fig. 1.9 Development and diversification of mechatronics technologies (after [17])

Fig. 1.10 The challenges of innovation

In addition to the challenges to course design associated with developments in

Table 1.4 Chapter structure Chapter(s) Subject area

© Springer International Publishing Switzerland 2016 17

Fig. 2.2 Many mono-disciplinary solutions for a given problem [1]

2.2 Computer Controlled Devices

The performance improvements due to mechatronic thinking have been profound

Fig. 2.3 Wafer scanner

Fig. 2.4 The performance

Fig. 2.5 The Preceyes eye surgery robot [5]

2.6 Cyber-Physical Systems, Smart Industry

The shift from decentralized mechatronic systems towards networked con-

1.2.2 Privacy and Security

1.2.3 Complexity and Ethics

2.2 Computer Controlled Devices

2.6 Cyber-Physical Systems, Smart Industry

2.7 Towards Systems Integration

3.2 Challenges in Design—Architecture and Sizing

3.2.1 Using Models to Size an Architecture

3.2.2 Coupled Disciplines for the Design of Mechatronic

3.2.3 Ability to Synthesize Optimal Architectures

3.2.4 Study of Safety and Availability in Mechatronic

3.2.5 Functional Virtual Representation of the Product

3.3 Challenges in Verification and Validation

3.3.1 Ability to Virtually Validate a System

3.3.2 Formalize Model Requests to Model Suppliers

3.3.3 Ability to Exchange and Seamlessly Integrate Models

3.3.4 Formal Verification of a Mechatronic System Through

3.3.5 Ability to Optimize Test Campaigns

3.4.1 Ability to Use Design Models to Improve Maintenance

3.4.2 Ability to Use Design Models to Improve Control

4.1.1 Complex System Design

4.1.2 Complex Versus Conventional System Design

4.1.3 Methodology Adoption Resistance

4.3.2 Experiment 1—Natural Experiment

4.3.3 Experiment 2—Controlled Comparative Study