Download the Full Version of textbook for Fast Typing at textbookfull.

com

AETA 2018 - Recent Advances in Electrical

Engineering and Related Sciences: Theory and
Application Ivan Zelinka

https://textbookfull.com/product/aeta-2018-recent-advances-
in-electrical-engineering-and-related-sciences-theory-and-
application-ivan-zelinka/

OR CLICK BUTTON

DOWNLOAD NOW

Download More textbook Instantly Today - Get Yours Now at textbookfull.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

AETA 2019 - Recent Advances in Electrical Engineering and

Related Sciences: Theory and Application Dario Fernando
Cortes Tobar
https://textbookfull.com/product/aeta-2019-recent-advances-in-
electrical-engineering-and-related-sciences-theory-and-application-
dario-fernando-cortes-tobar/
textboxfull.com

AETA 2017 - Recent Advances in Electrical Engineering and

Related Sciences: Theory and Application 1st Edition Vo
Hoang Duy Et Al. (Eds.)
https://textbookfull.com/product/aeta-2017-recent-advances-in-
electrical-engineering-and-related-sciences-theory-and-
application-1st-edition-vo-hoang-duy-et-al-eds/
textboxfull.com

Recent Advances in Mathematics for Engineering

(Mathematical Engineering, Manufacturing, and Management
Sciences) 1st Edition Mangey Ram (Editor)
https://textbookfull.com/product/recent-advances-in-mathematics-for-
engineering-mathematical-engineering-manufacturing-and-management-
sciences-1st-edition-mangey-ram-editor/
textboxfull.com

Recent Advances in Traffic Engineering: Select Proceedings

of RATE 2018 Shriniwas S. Arkatkar

https://textbookfull.com/product/recent-advances-in-traffic-
engineering-select-proceedings-of-rate-2018-shriniwas-s-arkatkar/

textboxfull.com
Soft Computing Systems Second International Conference
ICSCS 2018 Kollam India April 19 20 2018 Revised Selected
Papers Ivan Zelinka
https://textbookfull.com/product/soft-computing-systems-second-
international-conference-icscs-2018-kollam-india-
april-19-20-2018-revised-selected-papers-ivan-zelinka/
textboxfull.com

Recent Advances in Operator Theory and Operator Algebras

1st Edition Bercovici

https://textbookfull.com/product/recent-advances-in-operator-theory-
and-operator-algebras-1st-edition-bercovici/

textboxfull.com

Advances in Informatics and Computing in Civil and

Construction Engineering: Proceedings of the 35th CIB W78
2018 Conference: IT in Design, Construction, and
Management Ivan Mutis
https://textbookfull.com/product/advances-in-informatics-and-
computing-in-civil-and-construction-engineering-proceedings-of-
the-35th-cib-w78-2018-conference-it-in-design-construction-and-
management-ivan-mutis/
textboxfull.com

Scientific Computing in Electrical Engineering: SCEE 2018,

Taormina, Italy, September 2018 Giuseppe Nicosia

https://textbookfull.com/product/scientific-computing-in-electrical-
engineering-scee-2018-taormina-italy-september-2018-giuseppe-nicosia/

textboxfull.com

Physical Sciences and Engineering Advances in Life

Sciences and Oncology A WTEC Global Assessment 1st Edition
Paul Janmey
https://textbookfull.com/product/physical-sciences-and-engineering-
advances-in-life-sciences-and-oncology-a-wtec-global-assessment-1st-
edition-paul-janmey/
textboxfull.com
Lecture Notes in Electrical Engineering 554

Ivan Zelinka
Pavel Brandstetter
Tran Trong Dao
Vo Hoang Duy
Sang Bong Kim Editors

AETA 2018 - Recent

Advances in Electrical
Engineering and
Related Sciences:
Theory and Application
Lecture Notes in Electrical Engineering

Volume 554

Series Editors
Leopoldo Angrisani, Department of Electrical and Information Technologies Engineering, University of Napoli
Federico II, Napoli, Italy
Marco Arteaga, Departament de Control y Robótica, Universidad Nacional Autónoma de México, Coyoacán,
Mexico
Bijaya Ketan Panigrahi, Electrical Engineering, Indian Institute of Technology Delhi, New Delhi, Delhi, India
Samarjit Chakraborty, Fakultät für Elektrotechnik und Informationstechnik, TU München, München, Germany
Jiming Chen, Zhejiang University, Hangzhou, Zhejiang, China
Shanben Chen, Materials Science & Engineering, Shanghai Jiao Tong University, Shanghai, China
Tan Kay Chen, Department of Electrical and Computer Engineering, National University of Singapore,
Singapore, Singapore
Rüdiger Dillmann, Humanoids and Intelligent Systems Lab, Karlsruhe Institute for Technology, Karlsruhe,
Baden-Württemberg, Germany
Haibin Duan, Beijing University of Aeronautics and Astronautics, Beijing, China
Gianluigi Ferrari, Università di Parma, Parma, Italy
Manuel Ferre, Centre for Automation and Robotics CAR (UPM-CSIC), Universidad Politécnica de Madrid,
Madrid, Madrid, Spain
Sandra Hirche, Department of Electrical Engineering and Information Science, Technische Universität
München, München, Germany
Faryar Jabbari, Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA,
USA
Limin Jia, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland
Alaa Khamis, German University in Egypt El Tagamoa El Khames, New Cairo City, Egypt
Torsten Kroeger, Stanford University, Stanford, CA, USA
Qilian Liang, Department of Electrical Engineering, University of Texas at Arlington, Arlington, TX, USA
Ferran Martin, Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelona, Bellaterra,
Barcelona, Spain
Tan Cher Ming, College of Engineering, Nanyang Technological University, Singapore, Singapore
Wolfgang Minker, Institute of Information Technology, University of Ulm, Ulm, Germany
Pradeep Misra, Department of Electrical Engineering, Wright State University, Dayton, OH, USA
Sebastian Möller, Quality and Usability Lab, TU Berlin, Berlin, Germany
Subhas Mukhopadhyay, School of Engineering & Advanced Technology, Massey University,
Palmerston North, Manawatu-Wanganui, New Zealand
Cun-Zheng Ning, Electrical Engineering, Arizona State University, Tempe, AZ, USA
Toyoaki Nishida, Graduate School of Informatics, Kyoto University, Kyoto, Kyoto, Japan
Federica Pascucci, Dipartimento di Ingegneria, Università degli Studi “Roma Tre”, Rome, Italy
Yong Qin, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China
Gan Woon Seng, School of Electrical & Electronic Engineering, Nanyang Technological University,
Singapore, Singapore
Joachim Speidel, Institute of Telecommunications, Universität Stuttgart, Stuttgart, Baden-Württemberg,
Germany
Germano Veiga, Campus da FEUP, INESC Porto, Porto, Portugal
Haitao Wu, Academy of Opto-electronics, Chinese Academy of Sciences, Beijing, China
Junjie James Zhang, Charlotte, NC, USA
The book series Lecture Notes in Electrical Engineering (LNEE) publishes the latest developments
in Electrical Engineering - quickly, informally and in high quality. While original research
reported in proceedings and monographs has traditionally formed the core of LNEE, we also
encourage authors to submit books devoted to supporting student education and professional
training in the various fields and applications areas of electrical engineering. The series cover
classical and emerging topics concerning:
• Communication Engineering, Information Theory and Networks
• Electronics Engineering and Microelectronics
• Signal, Image and Speech Processing
• Wireless and Mobile Communication
• Circuits and Systems
• Energy Systems, Power Electronics and Electrical Machines
• Electro-optical Engineering
• Instrumentation Engineering
• Avionics Engineering
• Control Systems
• Internet-of-Things and Cybersecurity
• Biomedical Devices, MEMS and NEMS

For general information about this book series, comments or suggestions, please contact leontina.
[email protected].
To submit a proposal or request further information, please contact the Publishing Editor in
your country:
China
Jasmine Dou, Associate Editor ([email protected])
India
Swati Meherishi, Executive Editor ([email protected])
Aninda Bose, Senior Editor ([email protected])
Japan
Takeyuki Yonezawa, Editorial Director ([email protected])
South Korea
Smith (Ahram) Chae, Editor ([email protected])
Southeast Asia
Ramesh Nath Premnath, Editor ([email protected])
USA, Canada:
Michael Luby, Senior Editor ([email protected])
All other Countries:
Leontina Di Cecco, Senior Editor ([email protected])
Christoph Baumann, Executive Editor ([email protected])

** Indexing: The books of this series are submitted to ISI Proceedings, EI-Compendex,
SCOPUS, MetaPress, Web of Science and Springerlink **

More information about this series at http://www.springer.com/series/7818

Ivan Zelinka Pavel Brandstetter
• •

Tran Trong Dao Vo Hoang Duy

• •

Sang Bong Kim

Editors

AETA 2018 - Recent

Advances in Electrical
Engineering
and Related Sciences:
Theory and Application

123
Editors
Ivan Zelinka Pavel Brandstetter
Department of Computer Science, Department of Electronics,
Faculty of Electrical Engineering Faculty of Electrical Engineering
VŠB-TUO VŠB-TUO
Ostrava-Poruba, Czech Republic Ostrava, Czech Republic

Tran Trong Dao Vo Hoang Duy

International Cooperation, Department of Automatic Control
Research and Training Institute Ton Duc Thang University
Ton Duc Thang University Ho Chi Minh City, Vietnam
Ho Chi Minh, Vietnam

Sang Bong Kim

Department of Mechanical Design
Engineering
Pukyong National University
Busan, Korea (Republic of)

ISSN 1876-1100 ISSN 1876-1119 (electronic)

Lecture Notes in Electrical Engineering
ISBN 978-3-030-14906-2 ISBN 978-3-030-14907-9 (eBook)
https://doi.org/10.1007/978-3-030-14907-9

Library of Congress Control Number: 2019933407

© Springer Nature Switzerland AG 2020

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. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Foreword

The modern world is based on vitally important technologies that merge electronics,
cybernetics, computer science, telecommunication, and physics together. Since the
beginning of our technologies, we have been confronted with numerous techno-
logical challenges such as finding the optimal solution of various problems
including controlling technologies, power sources construction, and robotics.
Technology development of those and related areas has had and continues to have
profound impact on our civilization and our future lifestyle.
Therefore, this proceeding book containing articles of international conference
AETA 2018 is a timely volume to be welcomed by the community focused on
telecommunication, power control, and optimization as well as computational
science community and beyond.
This proceeding book consists of the hottest topic areas of selected papers like
telecommunication, power systems, digital signal processing, robotics, control
system, renewable energy, power electronics, soft computing, and more. All
selected papers represent interesting ideas and state-of-the-art overview.
Participations were carefully selected and reviewed; hence, this proceeding book
certainly is one of the few discussing the benefit from the intersection of those
modern and fruitful scientific fields of research. We hope that the proceeding book
will be an instructional material for senior undergraduate and entry-level graduate
students working in the area of electronics, power technologies, energy distribution,
control and robotics, etc. The proceeding book will also be a resource and material
for practitioners who want to apply discussed topics to solve real-life problems in
their challenging applications. The important part of proceeding book is partici-
pation of four keynote speakers from the Russia, USA, and two from India.
The decision to organize AETA conference and to create this proceeding book
was based on facts that technologies mentioned above, their use, and impact on life
are interesting areas, which are under intensive research from many other branches
of science today. This proceeding book is written to contain simplified versions of

v
vi Foreword

experiments with the aim to show how, in principle, problems about power systems
can be solved.
It is obvious that this proceeding book does not encompass all aspects of
discussed topics due to limited space and time of the conference. Only the main
ideas and results of selected papers are reported here. The authors and editors hope
that the readers will be inspired to do their own experiments and simulations, based
on information reported in this proceeding book, thereby moving beyond the scope
of it.
This proceeding book is devoted to the studies of common and related subjects
in intensive research fields of modern electric, electronic, and related technologies.
For these reasons, we believe that this proceeding book will be useful for scientists
and engineers working in the above-mentioned fields of research and applications.
At the end, we would like to thank Ton Duc Thang University (Ho Chi Minh
City, Vietnam) and VŠB-Technical University (Ostrava, Czech Republic) for
interest and strong support in AETA conference organization. Also, many thanks
belong to Springer Publishing Company for its highly professional, precise, and
quick production process. Without all of this, it would be impossible to organize
successful conference joining participants from the whole world.
September 2018 Editors
Foreword vii

This conference was supported by the Ton Duc Thang University (Ho Chi Minh
City, Vietnam) and VŠB - Technical University (Ostrava, Czech Republic).
Contents

Computer Science
New Neuromorphic AI NM500 and Its ADAS Application . . . . . . . . . . 3
Jungyun Kim
Analyzing l1-loss and l2-loss Support Vector Machines Implemented
in PERMON Toolbox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Marek Pecha and David Horák
Hybrid Fuzzy Neural Model Based Dempster-Shafer System
for Processing of Diagnostic Information . . . . . . . . . . . . . . . . . . . . . . . . 24
Alexander I. Dolgiy, Sergey M. Kovalev, Andrey V. Sukhanov,
and Vitezslav Styskala
ANFIS and Fuzzy Tuning of PID Controller for STATCOM
to Enhance Power Quality in Multi-machine System
Under Large Disturbance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Huu Vinh Nguyen, Hung Nguyen, and Kim Hung Le
Proposal of Electrode System for Measuring Level of Glucose
in the Blood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Klara Fiedorova and Martin Augustynek
Substitution Rules with Respect to a Context . . . . . . . . . . . . . . . . . . . . . 55
Michal Fait and Marie Duží
Fuzzy Model Predictive Control for Discrete-Time System
with Input Delays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Sofiane Bououden, Ilyes Boulkaibet, Mohammed Chadli, and Ivan Zelinka
An Improvement of Fuzzy-Based Control Strategy for a Series
Hydraulic Hybrid Truck . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Tri-Vien Vu, Bach H. Dinh, Anh-Minh Duc Tran, Chih-Keng Chen,
and Trung-Hieu Vu

ix
x Contents

A New Approach Newton-Raphson Load Flow Analysis in Power

System Networks with STATCOM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Dung Vo Tien, Radomir Gono, and Zbigniew Leonowicz
Neural Network for Smart Adjustment of Industrial Camera - Study
of Deployed Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Petr Dolezel and Daniel Honc
Risk Assessment Approach to Estimate Security of Cryptographic
Keys in Quantum Cryptography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Marcin Niemiec, Miralem Mehic, and Miroslav Voznak
Wavelet Transform Decomposition for Fetal Phonocardiogram
Extraction from Composite Abdominal Signal . . . . . . . . . . . . . . . . . . . . 125
Radana Kahankova and Radek Martinek
A CUDA Approach for Scenario Reduction in Hedging Models . . . . . . 134
Donald Davendra, Chin-mei Chueh, and Emmanuel Hamel
Using a Strain Gauge Load Cell for Analysis of Round Punch . . . . . . . 144
Dora Lapkova
Geometrical Computational Method to Locate Hypocenter
by Signal Readings from a Three Receivers . . . . . . . . . . . . . . . . . . . . . . 154
Alexander D. Krutas, Tatyana A. Smaglichenko, Alexander Smaglichenko,
and Maria Sayankina
An Intelligent Question-Answer System over
Natural-Language Texts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Marie Duží and Bjørn Jespersen
An Efficient Reduced Basis Construction for Stochastic Galerkin
Matrix Equations Using Deflated Conjugate Gradients . . . . . . . . . . . . . 175
Michal Béreš
An Investigation on Signal Comparison by Measuring of Numerical
Strings Similarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Alexander Smaglichenko, Tatyana A. Smaglichenko, Arkady Genkin,
and Boris Melnikov

Optimization
A Lightweight SHADE-Based Algorithm for Global
Optimization - liteSHADE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Adam Viktorin, Roman Senkerik, Michal Pluhacek, Tomas Kadavy,
and Roman Jasek
Pupil Localization Using Self-organizing Migrating Algorithm . . . . . . . 207
Radovan Fusek and Petr Dobeš
Contents xi

Differential Evolution Algorithms Used to Optimize Weights of Neural

Network Solving Pole-Balancing Problem . . . . . . . . . . . . . . . . . . . . . . . 217
Jan Vargovsky and Lenka Skanderova
The Use of Radial Basis Function Surrogate Models for Sampling
Process Acceleration in Bayesian Inversion . . . . . . . . . . . . . . . . . . . . . . 228
Simona Domesová
An Optimised Hybrid Group Method in Data Handling
(GMDH) Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Donald Davendra and Petr Martinek
A Better Indexing Method for Closest Open Location Policy
in Forklift Warehouse Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Duy Anh Nguyen, Truong Thinh Pham, and Viet Duong Nguyen
On-Line Efficiency-Optimization Control of Induction Motor Drives
Using Particle Swarm Optimization Algorithm . . . . . . . . . . . . . . . . . . . 261
Sang Dang Ho, Pavel Brandstetter, Cuong Tran Dinh, Thinh Cong Tran,
Minh Chau Huu Nguyen, and Bach Hoang Dinh
Introducing the Run Support Strategy for the Bison Algorithm . . . . . . 272
Anezka Kazikova, Michal Pluhacek, Tomas Kadavy, and Roman Senkerik
Optimizing Automated Storage and Retrieval Algorithm in Cold
Warehouse by Combining Dynamic Routing and Continuous
Cluster Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Ngoc Cuong Truong, Truong Giang Dang, and Duy Anh Nguyen
Dependency of GPA-ES Algorithm Efficiency on ES Parameters
Optimization Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Tomas Brandejsky
A Modified Bat Algorithm to Improve the Search Performance
Applying for the Optimal Combined Heat and Power Generations . . . . 303
Bach H. Dinh, An H. Ngo, and Thang T. Nguyen
Prediction of Hourly Vehicle Flows by Optimized
Evolutionary Fuzzy Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Pavel Krömer, Jana Nowaková, Martin Hasal, and Jan Platoš
A New Simple, Fast and Robust Total Least Square Error
Computation in E2: Experimental Comparison . . . . . . . . . . . . . . . . . . . 325
Michal Smolik, Vaclav Skala, and Zuzana Majdisova
On the Self-organizing Migrating Algorithm Comparison
by Means of Centrality Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Lukas Tomaszek, Patrik Lycka, and Ivan Zelinka
xii Contents

A Brief Overview of the Synergy Between Metaheuristics

and Unconventional Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
Roman Senkerik

Telecommunications
An Examination of Outage Performance for Selected Relay and Fixed
Relay in Cognitive Radio-Aided NOMA . . . . . . . . . . . . . . . . . . . . . . . . . 359
Tam Nguyen Kieu, Hong Nhu Nguyen, Long Nguyen Ngoc,
Tu-Trinh Thi Nguyen, Jaroslav Zdralek, and Miroslav Voznak
Throughput Analysis of Power Beacon-Aided Multi-hop Relaying
Networks Employing Non-orthogonal Multiple Access
with Hardware Impairments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
Phu Tran Tin, Pham Minh Nam, Tran Trung Duy, Phuong T. Tran,
Tam Nguyen Kieu, and Miroslav Voznak
Optimum Selection of the Reference Signal for Correlation Receiver
Applied to Marker Localization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
Martin Vestenický and Peter Vestenický
Comparing of Transfer Process Data in PLC and MCU
Based on IoT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
Antonin Gavlas, Jiri Koziorek, and Robert Rakay
Protecting Gateway from ABP Replay Attack on LoRaWAN . . . . . . . . 400
Erik Gresak and Miroslav Voznak
Development of a Distributed VoIP Honeypot System with Advanced
Malicious Traffic Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
Ladislav Behan, Lukas Sevcik, and Miroslav Voznak
Proposal and Implementation of Probe for Sigfox Technology . . . . . . . . 420
Jakub Jalowiczor and Miroslav Voznak
IoT Approach to Street Lighting Control Using MQTT Protocol . . . . . 429
Radim Kuncicky, Jakub Kolarik, Lukas Soustek, Lumir Kuncicky,
and Radek Martinek

Materials
Temperature Dependence of Microstructure
in Liquid Aluminosilicate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
Mai Van Dung, Le The Vinh, Vo Hoang Duy, Nguyen Kieu Tam,
Tran Thanh Nam, Nguyen Manh Tuan, Truong Duc Quynh,
and Nguyen Van Yen
Contents xiii

Study on Effect of Parameters on Friction Stir Welding Process

of 6061 Aluminum Alloy Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
Van Vu Nguyen, Hoang Linh Nguyen, Tan Tien Nguyen,
Thien Phuc Tran, and Sang Bong Kim
Convergence Study of Different Approaches of Solving
the Hartree-Fock Equation on the Potential Curve
of the Hydrogen Fluoride . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461
Martin Mrovec

Control Systems
Network Traffic Anomaly Detection in Railway Intelligent Control
Systems Using Nonlinear Dynamics Approach . . . . . . . . . . . . . . . . . . . . 475
Maria A. Butakova, Andrey V. Chernov, Sergey M. Kovalev,
Andrey V. Sukhanov, and Stanislav Zajaczek
Advanced Methods of Detection of the Steganography Content . . . . . . . 484
Jakub Hendrych and Lačezar Ličev
Robust Servo Controller Design Based on Linear Shift Invariant
Differential Operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494
Dae Hwan Kim and Sang Bong Kim
Servo Controller Design and Fault Detection Algorithm for Speed
Control of a Conveyor System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
Trong Hai Nguyen, Nguyen Thanh Phuong, and Hung Nguyen
A Control System for Power Electronics with an NXP Kinetis
Series Microcontroller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514
Daniel Kouřil, Martin Sobek, and Petr Chamrád
A MIMO Robust Servo Controller Design Method
for Omnidirectional Automated Guided Vehicles Using Polynomial
Differential Operator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
Van Lanh Nguyen, Sung Won Kim, Choong Hwan Lee, Dae Hwan Kim,
Hak Kyeong Kim, and Sang Bong Kim
Model Reference Adaptive Control Strategy for Application
to Robot Manipulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533
Manh Son Tran, Suk Ho Jung, Nhat Binh Le, Huy Hung Nguyen,
Dac Chi Dang, Anh Minh Duc Tran, and Young Bok Kim
Stabilization of Time-Varying Systems Subject to Actuator
Saturation: A Takagi-Sugeno Approach . . . . . . . . . . . . . . . . . . . . . . . . . 548
Sabrina Aouaouda and Mohammed Chadli
xiv Contents

Observer Based Control for Systems with Mismatched Uncertainties

in Output Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
Van Van Huynh, Tran Thanh Phong, and Bach Hoang Dinh
Nonlinear Disturbance Observer with Recurrent Neural
Network Compensator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569
Shihono Yamada and Jun Ishikawa
Parameters Estimation for Sensorless Control of Induction Motor
Drive Using Modify GA and CSA Algorithm . . . . . . . . . . . . . . . . . . . . . 580
Thinh Cong Tran, Pavel Brandstetter, Cuong Dinh Tran, Sang Dang Ho,
Minh Chau Huu Nguyen, and Pham Nhat Phuong
Study on Algorithms and Path-Optimization for USV’s
Obstacle Avoidance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592
Ngoc-Huy Tran and Nguyen Nhut-Thanh Pham
Visual Servoing Controller Design Based on Barrier Lyapunov
Function for a Picking System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 605
Jong Min Oh, Jotje Rantung, Sung Rak Kim, Sang Kwun Jeong,
Hak Kyeong Kim, Sea June Oh, and Sang Bong Kim
Designing a PID Controller for Ship Autopilot System . . . . . . . . . . . . . 618
Dinh Due Vo, Viet Anh Pham, Phung Hung Nguyen,
and Duy Anh Nguyen
The Rotor Initial Position Determination of the Hi-Speed
Switch-Reluctance Electrical Generator for the Steam-Microturbine . . . 628
Pavel G. Kolpakhchyan, Vladimir I. Parshukov, Boris N. Lobov,
Nikolay N. Efimov, and Vadim V. Kopitza
Stability and Chaotic Attractors of Memristor-Based Circuit
with a Line of Equilibria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 639
N. V. Kuznetsov, T. N. Mokaev, E. V. Kudryashova, O. A. Kuznetsova,
R. N. Mokaev, M. V. Yuldashev, and R. V. Yuldashev

Mechanical Engineering
Behavior of Five-Pad Tilting–Pad Journal Bearings with Different
Pivot Stiffness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647
Phuoc Vinh Dang, Steven Chatterton, and Paolo Pennacchi
Dynamic Characteristics of a Non-symmetric Tilting Pad
Journal Bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 658
Phuoc Vinh Dang, Steven Chatterton, and Paolo Pennacchi
Contents xv

Energy
DCM Boost Converter in CPM Operation for Tuning Piezoelectric
Energy Harvesters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 673
Andrés Gomez-Casseres, David Florez, and Darío Cortes
Effect of Weighting Coefficients on Behavior of the DTC Method
with Direct Calculation of Voltage Vector . . . . . . . . . . . . . . . . . . . . . . . 683
Jakub Baca, Martin Kuchar, and Petr Palacky
A New Protocol for Energy Harvesting Decode-and-Forward
Relaying Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693
Duy-Hung Ha, Dac-Binh Ha, Jaroslav Zdralek, Miroslav Voznak,
and Tan N. Nguyen
Average Bit Error Probability Analysis for Cooperative DF Relaying
in Wireless Energy Harvesting Networks . . . . . . . . . . . . . . . . . . . . . . . . 705
Hoang-Sy Nguyen, Thanh-Sang Nguyen, Tan N. Nguyen,
and Miroslav Voznak
LCCT vs. LLC Converter - Analysis of Operational Characteristics
During Critical Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715
Michal Pridala, Michal Frivaldsky, and Pavol Spanik
Control Renewable Energy System and Optimize Performance
by Using Weather Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 725
Duy Tan Nguyen, Duy Anh Nguyen, and Lien Son Chau Hoang
Analysis of Efficiency and THD in 7-Level Voltage Inverters
with Reduced Number of Switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 736
Ales Havel, Martin Sobek, and Petr Chamrad
Waste Management - Weighing-Machine Automation . . . . . . . . . . . . . . 747
Zdenek Slanina, Rostislav Pokorny, and Jan Dedek
Optimization of Voltage Model for MRAS Based Sensorless Control
of Induction Motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 758
Ondrej Lipcak and Jan Bauer
Capability of Predictive Torque Control Method to Control
DC-Link Voltage Level in Small Autonomous Power System
with Induction Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769
Pavel Karlovsky and Jiri Lettl
Feasibility Structural Analysis of Engineering Plastic Reel Module
for Carrying Wound High-Voltage Electric Transmission Line . . . . . . . 778
Jungyun Kim, Ho-Young Kang, Young-Geon Song, and Chan-Jung Kim
xvi Contents

Improving Fault Tolerant Control to the One Current Sensor Failures

for Induction Motor Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789
Cuong Dinh Tran, Pavel Brandstetter, Sang Dang Ho, Thinh Cong Tran,
Minh Chau Huu Nguyen, Huy Xuan Phan, and Bach Hoang Dinh
Impact of Parameter Variation on Sensorless Indirect Field Oriented
Control of Induction Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 799
Andrej Kacenka, Pavol Makys, and Lubos Struharnansky
Validation the FEM Model of Asynchronous Motor by Analysis
of External Radial Stray Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 810
Petr Kacor and Petr Bernat
Outage and Intercept Probability Analysis for Energy-Harvesting-
Based Half-Duplex Relay Networks Assisted by Power Beacon Under
the Existence of Eavesdropper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 821
Tan N. Nguyen, Phuong T. Tran, Nguyen Dao, and Miroslav Voznak
Design of Electrical Regulated Drainage with Energy Harvesting . . . . . 835
Vaclav Kolar, Roman Hrbac, Tomas Mlcak, and Jiri Placek
Analysis of Appliance Impact on Total Harmonic Distortion
in Off-Grid System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844
Michal Petružela, Vojtěch Blažek, and Jan Vysocký
Influencing of Current Sensors by an External Magnetic Field
of a Nearby Busbar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 850
Tadeusz Sikora and Jan Hurta
A Model for Predicting Energy Savings Attainable by Using Lighting
Systems Dimmable to a Constant Illuminance Level . . . . . . . . . . . . . . . 860
T. Novak, J. Sumpich, J. Vanus, K. Sokansky, R. Gono, J. Latal,
and P. Valicek
Strategy of Metropolis Electrical Energy Supply . . . . . . . . . . . . . . . . . . 870
Valery Beley, Andrey Nikishin, and Dmitriy Gorbatov

Robotics
Attitude Control of Jumping Robot with Bending-Stretching
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 883
Chea Xin Ong, Yurika Nomura, and Jun Ishikawa
Geometric Foot Location Determination Algorithm
for Façade Cleaning Robot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 894
Shunsuke Nansai and Hiroshi Itoh
Contents xvii

Smart Manipulation Approach for Assistant Robot . . . . . . . . . . . . . . . . 904

Yeyson Becerra, Jaime Leon, Santiago Orjuela, Mario Arbulu,
Fernando Matinez, and Fredy Martinez
Computational Study on Upward Force Generation of Gymnotiform
Undulating Fin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 914
Van Hien Nguyen, Canh An Tien Pham, Van Dong Nguyen,
Hoang Long Phan, and Tan Tien Nguyen
Modular Design of Gymnotiform Undulating Fin . . . . . . . . . . . . . . . . . 924
Van Dong Nguyen, Canh An Tien Pham, Van Hien Nguyen,
Thien Phuc Tran, and Tan Tien Nguyen
Path Following Control of Automated Guide Vehicle
Using Camera Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 932
Dae Hwan Kim and Sang Bong Kim
Binary Classification of Terrains Using Energy Consumption
of Hexapod Robots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 939
Valeriia Iegorova and Sebastián Basterrech
The Movement of Swarm Robots in an Unknown
Complex Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 949
Quoc Bao Diep and Ivan Zelinka

Image Processing
Contour Detection Method of 3D Fish Using a Local Kernel
Regression Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 963
Jong Min Oh, Sung Rak Kim, Sung Won Kim, Nam Soo Jeong,
Min Saeng Shin, Hak Kyeong Kim, and Sang Bong Kim
Camera Based Tests of Dimensions, Shapes and Presence Based
on Virtual Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 973
Lukas Soustek, Radek Martinek, Lukas Snajdr, and Petr Bilik
A 3D Scanner Based on Virtual Instrumentation Implemented by a 1D
Laser Triangulation Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 982
Jindrich Brablik, Radek Martinek, Marek Haluska, and Petr Bilík
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 991
Computer Science
 New Neuromorphic AI NM500
and Its ADAS Application

Jungyun Kim(&)

School of Mechanical and Automotive Engineering,

Catholic University of Daegu, 13-13, Hayang-ro, Gyeongsan-si,
Gyeongsangbuk-do 38430, Korea
[email protected]

Abstract. This article deals with an ADAS (Advanced Driver Assistance

System) application using newly developed neuromorphic artificial intelligent
chip NM500. Neuromorphic artificial intelligence is distinguished from other AI
by its particular hardware structure and parallel algorithms of learning and
recognition. Thus, neurons of NM500 can learn and recognize patterns extracted
from any data sources with less energy and complexity than modern micro-
processors. The proposed application can control the vehicle speed by recog-
nizing the traffic information images marked on road. We have built a small-
scaled vehicle model to discuss the real-time performance as well as hardware
implementation with NM500. Taking advantages of NM500, the system simply
consists of a low-priced surveillance camera attached in the front windshield of a
vehicle and an Arduino kit, which processes the video signal from the camera
and speed control signal.

Keywords: NM500
Neuromorphic chip
Artificial intelligence

Edge computing
ADAS application

1 Introduction

Last year, Nepes, a highly reliable semiconductor manufacturer in Korea, has

announced the newly developed neuromorphic artificial intelligent chip named NM500
and NeuroShield: a trainable board for IoT and smart appliances featuring NM500
chip, which can interface with Arduino boards or a PC (Fig. 1). NM500 is a hardware
artificial intelligence opening new frontiers for smart sensors, IoT (Internet of Things),
machine learning and cognitive computing. Its neurons can learn and recognize pat-
terns extracted from any data sources such as images, audio waveform, bio signals, text
and more, with less energy and complexity than modern microprocessors.
Furthermore, NM500 is a product that enables Edge computing, an extension of
IoT, which is currently spreading throughout the industry. Edge computing is a
machine that judges a machine to suit each situation and takes action accordingly [1].
While IoT provides information about each situation to the user and actions taken
according to the judgment of the user, Edge computing judges and learns on its own,
and judges appropriate situation. In short, each machine is equipped with artificial
intelligence.

© Springer Nature Switzerland AG 2020

I. Zelinka et al. (Eds.): AETA 2018, LNEE 554, pp. 3–12, 2020.
https://doi.org/10.1007/978-3-030-14907-9_1
4 J. Kim

Neuromorphic computing has emerged in recent years as a complementary archi-

tecture to von Neumann systems. The term neuromorphic computing has been come out
1990 by Carver Mead [2]. This biologically inspired approach has created highly
connected synthetic neurons and synapses that can be used to model neuroscience
theories as well as solve machine-learning problems. The promise of the technology is
to create a brain-like ability to learn and adapt, and is notable for being highly connected
and parallel, requiring low-power, and collocating memory and processing [3, 4].
Many people think artificial intelligence in terms of software. As a representative
example, Alpha Go and Alexa use ‘deep learning’ technology which learn through
experience. These artificial intelligences require a large-capacity server storing data and
a supercomputer processing the data. However, neurons of neuromorphic chip can
learn and recognize patterns extracted from any data sources with less energy and
complexity than modern microprocessors. Thus, it can make decisions based on situ-
ations without a large server or supercomputer, and integrate artificial intelligence into
various industries.

Fig. 1. NM500 and NeuroShield.

In this paper, we have described an ADAS application using newly developed

neuromorphic chip NM500. The proposed application can control the vehicle speed by
recognizing the traffic information images marked on road such as crosswalk, school
zone, and road-bump, etc. First, we briefly reviewed the neuromorphic artificial
intelligence regarding its special features, neural network structure, and distinguished
algorithms of learning and recognition. Then the hardware architecture of NM500 was
introduced focused on the neuron connections and data flows. Finally, we have built a
small-scaled vehicle model to discuss the real-time performance as well as hardware
implementation with NM500. The system consists of a camera, a NeuroShield, and an
Arduino kit, which processes the video signal from the camera and speed control
signal.
 New Neuromorphic AI NM500 and Its ADAS Application 5

2 Neuromorphic Artificial Intelligence

Much of work in neuromorphic computing have been driven by the development of

hardware that could perform parallel operations, inspired by observed parallelism in
biological brains, but on a single chip. The popular motivations have been inherent
parallelism, extremely low power operation, real-time performance, speed in both
operation and training, and scalability. Because of their fault tolerance characteristics or
reliability in the face of hardware errors, neuromorphic computing has been imple-
mented in neural network-style architectures (i.e., architectures made up of neuron and
synapse-like components). Moreover, many neuromorphic systems perform learning
tasks in an unsupervised, low-power manner that is called as on-line learning defined as
the ability to adapt to changes in a task as they occur.
Neural network supported by NM500 is FFNN (Feed Forward Neural Network)
using RBF (radial basis function). It has a 3-layer logical structure with one input layer,
one hidden layer, and one output layer (Fig. 2). The hidden layer can be extended to the
number of physical neurons supported by NM500. For a single chip, 576 neurons are
available and can scale up to several hundred million with the same processing time
[5]. Each neuron in hidden layer has 256 bytes RAM and consists synapses. The
neurons are capable of ranking similarities between input vectors and the reference
patterns they hold in memory, but also reporting conflicting responses or cases of
uncertainty, reporting unknown responses or cases of anomaly or novelty.
NM500 calculates two norm values using L1 (Manhattan) or Lsup (Supremum)
method as a criterion for determining the similarity between learned data and input data
in a neuron. The recognition modes supported by NM500 provide RBF and KNN
(k-Nearest Neighbor) algorithms [6, 7]. The neurons can learn and recognize input
vectors autonomously and in parallel. If several neurons recognize a pattern, i.e.
“firing”, their responses can be retrieved automatically in increasing order of distance
(equivalent to a decreasing order of confidence). The information from a firing neuron
includes its distance, category, and neuron identifier. If the response is polled, this data
can be consolidated to make a more sophisticated decision weighing the cost of
uncertainty or else.
Learning is initiated by simply broadcasting a category, which is a label identifying
the data to be learned, after an input pattern. If it represents novelty, the next neuron
available in the chain automatically stores the pattern and its category. If some firing
neurons recognize the pattern but with a category other than the category to learn, they
auto-correct their influence fields. As the NM500 network broadcasts a new input
pattern, all the neurons update their distance simultaneously by using parallel bus
communication. They are ready to respond to a query as soon as the last component
received. The neurons reacting to an input pattern autonomously order themselves per
decreasing confidence. This unique feature pertains to the parallel architecture of
NM500 network, which allows a winner-takes-all among the reacting neurons.
6 J. Kim

Fig. 2. 3-layer FFNN in NM500.

Table 1. Electrical and I/Os specifications of NM500.

Part Specification
Clock frequency 37 MHz for single chip
18 MHz for chain of multiple chips
I/O Parallel bus (26lines)
Electrical 3.3 V I/O operation, 1.2 V core
Power consumption < 153mWatt in active mode at 1.2 V and 3.3 V
Package 64-pin CSP 4.5  4.5  0.5 mm package

3 NM500 Architecture

The NM500 chip is composed of the following modules: neuron interconnect part and
chain of neurons: daisy-chained and interconnected (Fig. 3). Inter-module and inter-
neuron communications are made through a bi-directional parallel bus. The neuron cell
is composed of a memory and a set of six. The detailed specifications of NM500 are
listed in Table 1.
A neuron can have three states in the chain: Idle, RTL(Ready-To-Learn) or
Committed. It becomes committed as soon as it learns a pattern and its category register
is written with a value different from zero. Its DCO(Daisy-Chain-Out) control line
automatically rises, changing its status from RTL to Committed. The next neuron in the
chain becomes the RTL. It has its DCI(Daisy-Chain-In) high and DCO low. The
transfer of the DCI-DCO from one neuron to the next is activated in the same way
whether the two consecutive neurons belong to a same cluster or not.
One of the benefits of the NM500 architecture is that one can cascade multiple
chips in parallel to expand the size of the neural network by increment of 576 neurons.
The behavior of the neurons in a single-chip or multiple-chips configuration remains
the same.
 New Neuromorphic AI NM500 and Its ADAS Application 7

Fig. 3. NM500 architecture.

4 ADAS Application Using Image Learning and Recognition

Among various applications of NM500, we deal with an ADAS (Advanced Driver

Assistance System) application using image learning and recognition function. It works
like one of conventional ADAS functions such as cruise control or intelligent speed
assist system in passenger cars. It can control the vehicle speed by recognizing the
traffic information images marked on road such as crosswalk, school zone, and road-
bump, etc. The proposed system is motivated by making cheaper and simplifying the
current system (Fig. 4). Taking advantages of NM500, it consists of a low-priced
surveillance camera attached in the front windshield of a vehicle, an NM500, and an
Arduino kit, which processes the video signal from the camera and speed control
signal.

4.1 Learning and Recognition of Traffic Information Images

There have been a myriad of popular theoretic and practical results on the image-
processing field including learning and recognition solutions. In NM500 applications,
there also need typical image processing steps like other computer vision applications
(Fig. 5) [8–12]. Fortunately, one can carry out these pre-processing, learning, and
recognition steps easily by using Knowledge Studio: a novel development tool supplied
by Nepes. Though not skilled in a machine learning technology, one can develop
various AI applications with NM500 by using the built-in functions only.
Knowledge Studio is a GUI (Graphic User Interface) and Show-and-Tell based
development environment program. It can pre-process the data from various sensors
and extract the genetic feature of learned data by using selective algorithms of sub-
sampling, histogram, and composite profile. And it can decide, e.g., classification,
detection, and segmentation by interacting with neurons in NM500. Moreover, there
has open-source libraries of Arduino kit, thus it can communicate with external sensors
and actuators as like NM500 artificial intelligence works on-line.
In this study, we consider crosswalk and bump signs marked on the road for the
traffic information images. First, a video of 150 s has been recorded using a surveil-
lance camera in good driving conditions: bright sunshine weather in the afternoon,
average vehicle speed about 40 km/hr. Image pre-processing through the sub-sampling
algorithm has been carried out in average gray scale mode with Knowledge Studio
8 J. Kim

(a) Comparison of the proposed method with conventional ADAS application.

(b) Proposed ADAS algorithm.

Fig. 4. Structure and algorithm of proposed system.

Fig. 5. Image processing steps.

(Fig. 6(a)). The test videos have been recorded with the resolution of 1024  768
pixels and frame rates of 30 frames/sec. The ROI (Region of Interest) for learning and
recognition of road signs has been set to 270  70 pixels on the center of image area.
On the completing the training, only three neurons participate in learning the road
signs; one for the road-bump and two for the crosswalk (Fig. 6(b)). Because of their
relatively simple features, the learning and recognition of road signs can be accom-
plished even with a small number of neurons in NM500. Moreover, once new neurons
are learned, previously learned neurons coordinate themselves by interacting with each
other. Thus if the features of input data are similar to those of previously learned
neurons, the learning does not proceed since it can be judged by already learned
neurons. The evaluation of recognition performance has been resulted 88% using the
test video; 22 signs have been detected successfully from all 25 signs.
 New Neuromorphic AI NM500 and Its ADAS Application 9

(a) Learning and pre-processing of crosswalk image. (b) Resulted learning models.

Fig. 6. Learning and recognition with Knowledge Studio.

4.2 Real-Time Performance and Hardware Implementation

In order to discuss whether it is applicable to an ADAS application, the proposed
system must guarantee the real-time performance as well as hardware implementation
with NM500. Since it is beyond the scope of this study to build a test vehicle, we have
made a small-scaled vehicle model with dc motors, which can emulate the vehicle
movements.
The vehicle model is designed to validate the learning and recognition performance
in real-time. Therefore, it consists of a camera, an image processor, and a NeuroShield
(Fig. 7). And a micro-controller (Arduino Mega board) is added to manage the data and
control signals with motor driver and NM500. Arduino is an open-source electronics
platform based on easy-to-use hardware and software [9]. The camera and image
processor are chosen among available Arduino peripherals; the detailed specifications
of components are shown in Table 2.

(a) Rear view. (b) Front view.

Fig. 7. Small-scaled vehicle model with the proposed system.

10 J. Kim

Table 2. Specifications of the prototype components.

Part Specification
OV2640
2Mega pixels: 1600x1200 UXGA
Camera
1/4” sensor size
15~60 fps image transfer rate
3.2 inch TFT LCD for ROI set
Image processor SPI speed: 8MHz
Resolution support: 0.3MP ~ 5MP
Half-bridged DC motor controller
Motor driver 2 I/O channels
Available motor driving voltage: 4.7~24V
DC motors 5V DC, Installed a reduction gear
Arduino ATmega2560
I/O: Digital 54, Analog input 16
Micro-controller
Clock speed: 16MHz
Flash memory: 256 KB

The test consists of following procedures. First, NM500 in NeuroShield has learned
the road-bump sign by using Knowledge Studio or the embedded camera and image
processor (Fig. 8). Only one neuron of NM500 has been found to learn and detect this
image throughout the test. While moving on the flat road in constant speed (about
0.6 m/s), the miniature vehicle slows down its speed to 0.2 m/s, immediately on
detecting the sign (Fig. 9). From the repeated test results, NM500 finished the learning
process and surely detected the road sign in every case even with minimal number of
neurons. After recognizing the road-bump image, it took almost 0.5 s to the target
speed; it is short enough to guarantee the real-time performance in a real vehicle
application (Fig. 10).

(a) Learning with Knowledge Studio. (b) Learning using embedded camera.

Fig. 8. Learning the road bump sign.

 New Neuromorphic AI NM500 and Its ADAS Application 11

Fig. 9. Miniature test: the miniature vehicle, road, and road-bump.

0.7

0.6

Test #1
Wheel speed [m/sec]

0.5
Test #2
Test #3
0.4 Test #4
Test #5
0.3

0.2

0.1

0.0
2 3 4
Time [sec]

Fig. 10. Test results.

5 Conclusions

In this paper, we have described an ADAS application using newly developed neu-
romorphic chip NM500. Its neurons can learn and recognize patterns extracted from
any data sources with less energy and complexity than modern microprocessors. The
proposed application can control the vehicle speed by recognizing the traffic infor-
mation images marked on road. Taking advantages of NM500, the system consists of a
low-priced surveillance camera attached in the front windshield of a vehicle, an
NM500, and an Arduino kit, which processes the video signal from the camera and
speed control signal. Finally, in order to discuss the real-time performance as well as
hardware implementation, we have made a small-scaled vehicle model. From the
feasibility test results, NM500 finished the learning process and surely detected the
road sign in every case even with minimal number of neurons. The major conclusions
of this paper are as follows.
• NM500 is a hardware artificial intelligence and its neurons can learn and recognize
patterns extracted from any data sources. Neuromorphic computing has been
characterized by inherent parallelism, extremely low power operation, real-time
performance, speed in both operation and training, and scalability.
12 J. Kim

• The proposed ADAS application using image learning and recognition function of
NM500 can control the vehicle speed by recognizing the traffic information images
marked on road such as crosswalk, school zone, and road-bump, etc. Motivated by
making cheaper and simplifying the current system, it consists of a low-priced
surveillance camera attached in the front windshield of a vehicle, an NM500, and an
Arduino kit.
• In case of learning the crosswalk and bump signs marked on the road in bright
sunshine weather, three neurons participate in learning the road signs; one for the
road-bump and two for the crosswalk. And the evaluation of recognition perfor-
mance has been resulted 88% using the test video.
• A small-scaled vehicle model has been built using NM500 and Arduino peripherals
in order to discuss the real-time performance as well as hardware implementation.
From the feasibility test results, NM500 finished the learning process and surely
detected the road sign in every case even with minimal number of neurons.

Acknowledgement. This work was supported by research grants from the Catholic University
of Daegu in 2017.

References
1. Ai, Y., Peng, M., Zhang, K.: Edge computing technologies for internet of things: a primer.
Digit. Commun. Netw. 4(2), 77–86 (2018)
2. Mead, C.: Neuromorphic electronic systems. Proc. IEEE 78(10), 1629–1636 (1990)
3. Schuman, C.D., Potok, T.E., Patton, R.M., Birdwell, D., Dean, M.E., Rose, G.S., Plank, J.S.:
A Survey of Neuromorphic Computing and Neural Networks in Hardware, ArXiv:
1705.06963, pp. 1–88 (2017)
4. Gerven, M.: Computational foundations of natural intelligence. Front. Comput. Neurosci.
11, 112–136 (2017)
5. General Vision, NeuroMem Technology Reference Guide (2018)
6. Chen, S., Cowan, C.F.N., Grant, P.M.: Orthogonal least squares learning algorithm for radial
basis function networks. IEEE Trans. Neural Networks 2(2) (1991)
7. Keller, J.M., Gray, M.R., Givens Jr., J.A.: A fuzzy K-nearest neighbor algorithm. IEEE
Trans. Syst. Man Cybern. 4, 580–585 (1985). SMC-15
8. Sardar, S., Babu, K.A.: Hardware implementation of real-time, high performance, RCE-NN
based face recognition system. In: 27th International Conference on VLSI Design and 13th
International Conference on Embedded Systems, pp. 174–179 (2014)
9. Sinha, P., Balas, B., Ostrovsky, Y., Russell, R.: Face recognition by humans: nineteen results
all computer vision researchers should know about. Proc. IEEE 94(11) (2006)
10. Sardar, S., Tewari, G., Babu, K.A.: A hardware/software codesign model for face
recognition using cognimem neural network chip. In: International Conference on Image
Information Processing (ICIIP 2011) (2011)
11. Sui, C., Kwok, N.M., Ren, T.: A restricted coulomb energy (RCE) neural network system for
hand image segmentation. In: 2011 Canadian Conference on Computer and Robot Vision
(2011)
12. Batur, A.U., Flinchbaugh, B.E.: Performance analysis of face recognition algorithms on
tms320c64x, Texas Instruments Application Report-SPRA874, pp. 1–12 (2002)
 Analyzing l1-loss and l2-loss Support
Vector Machines Implemented
in PERMON Toolbox

Marek Pecha1,2(B) and David Horák1,2

1
Department of Applied Mathematics, VŠB – Technical University of Ostrava,
Ostrava, Czech Republic
[email protected]
2
Institute of Geonics CAS, Ostrava, Czech Republic

Abstract. This paper deals with investigating l1-loss and l2-loss

l2-regularized Support Vector Machines implemented in PermonSVM
– a part of our PERMON toolbox. The loss functions quantify error
between predicted and correct classifications of samples in cases of non-
perfectly linearly separable classifications. In numerical experiments, we
study properties of Hessians related to performance score of models and
analyze convergence rate on 4 public available datasets. The Modified
Proportioning and Reduced Gradient Projection algorithm is used as
a solver for the dual Quadratic Programming problem resulting from
Support Vector Machines formulations.

Keywords: Support Vector Machines · SVM · PermonSVM ·

Hinge loss functions · Quadratic Programming · QP · MPRGP

1 Support Vector Machines for Classifications

In the last two decades, the Support Vector Machines (SVMs) [4], due to
their accuracy and obliviousness to dimensionality [15], have become a popu-
lar machine learning technique with applications including genetics [3], image
processing [7], and weather forecasting [14]. In this paper, we are only interested
in SVMs for classification.
SVM is originally designed as a supervised binary classifier, i.e. a classifier
that decides whether a sample falls into either Class A (label 1) or Class B
(label −1) by means of a model. The model is determined from the already cat-
egorised training samples in the training phase of the classifier. Unless otherwise
stated, let us assume the training samples are linearly separable. The essential
idea of the SVM classifier training is to find the maximal-margin hyperplane that
divides the Class A from the Class B samples by the widest possible empty strip,
which is called the functional margin. The samples contributing to the definition
of such hyperplane are called the support vectors – see the circled samples lying
on the dashed hyperplanes depicted in Fig. 1.
c Springer Nature Switzerland AG 2020
I. Zelinka et al. (Eds.): AETA 2018, LNEE 554, pp. 13–23, 2020.
https://doi.org/10.1007/978-3-030-14907-9_2
Exploring the Variety of Random
Documents with Different Content
 Characteristics.

13. AN ANGLE BAY IN BRIDGE

STREET.
W. & S. OWEN,
Architects.

One thing which is at once obvious from the general scheme is the
adoption of open spaces, communal gardens, and allotments in
preference to the spaces which are devoted to individual gardens
surrounding each cottage in so many other places. There is
something to be said for and against this. The general amenity of
the village gains by the Port Sunlight method, whilst the special
charm of individual gardens which enthusiastic efforts produce is
naturally lacking. In this way we get less value of contrasts, and lose
something of that spirit of emulation which spurs the individual to
special effort. Of one thing, however, there can be no doubt. The
absence of the many dividing lines of fences between each cottage
frontage produces a breadth of effect along the lines of roadways
which is in itself very pleasing. From the point of view of the town-
planner who looks for the collective result this is, of course, very
satisfactory.
DOUGLAS AND FORDHAM, 14. SOME PARK ROAD HOUSES.
Architects.
 15. COTTAGES IN NEW CHESTER
ROAD.
W. OWEN, Architect.

Another thing which will be noticed in the illustrations is the

elevation of many of the houses above the level of the roadway. This
gives a much wider and pleasanter outlook from the windows of the
cottages, besides producing a much better effect in the buildings
from the roadway than when they are placed on the same level. The
sloping green banks leading up to terraced paths in front of the
cottages are a distinctive feature of the village. (See Pl. 4.)
It has been maintained that without a good deal of monotony you
cannot get very fine architectural results, and it must be admitted
that many examples go to prove it. There is a large surface of
monotony in the Pyramids; there is a marvellous monotony of detail
in the Houses of Parliament; there is a boundless monotony in the
house fronts in Gower Street, yet all these have been admired. So
this line of argument might have suggested the continued
employment of only one architect, or at least only one type of
design, for the cottages at Port Sunlight. The great variety of designs
in the cottages, which has proved one of the attractions of the place,
has, however, in some sense at least, justified itself. Even the
flamboyant Gothic dormers and the stepped Belgian gables have a
reacting influence on some of their neighbours, though we might
consider the latter rather unpractical on the one hand, or the former
too pretentious on the other. Moreover, whilst we wonder at the
generosity of view which could bestow some of these solid oak-
framed structures with their wealth of carving and enriched plaster
panellings on the working classes of an industrial village, we cannot
but feel grateful to the hand that gave them, though we ourselves
may never be able to afford such luxuries of the building art for
ourselves. May we not accept these as symbols of some kindly
gratitude with which a profitable company decorates the homes of
its industrial population? Honestly, we cannot regret these bonnes
bouches in the building scheme, though they bravely put out of sight
the counting-house and the rates of interest! These are really very
welcome ebullitions from that solid undercurrent of practical
economy which has placed the whole concern on a sound business
footing.
 16. GROUP AT ANGLE OF LOWER ROAD AND CENTRAL ROAD.
J. L. SIMPSON, Architect.
ERNEST GEORGE AND YEATES,
Architects.

17. A RECESSED GROUP IN GREENDALE ROAD.

C. H. REILLY,
Architect.

18. COTTAGES ON SEMI-CIRCULAR PLAN IN LOWER ROAD.

W. OWEN, Architect. 19. FIRST COTTAGES BUILT AT PORT SUNLIGHT.
Awarded Grand Prix, Brussels Exhibition, 1910, for their reproduction there.
GRAYSON AND OULD,
Architects.

20. A THREE-GABLED GROUP IN NEW CHESTER ROAD.

This element of variety which is so marked in the design of the

cottages at Port Sunlight has been obtained without much departure
from the genuine English type. Even where a Dutch or Belgian
character appears it is carried out with something of the breadth and
simplicity which one associates with purely English work. There is
very little, if anything, that could be called freakish or odd. The
stepped gables or the flamboyant dormers which vary the treatment
are not unacceptable as variants. As to the use of oak framing with
plaster panels—the familiar Old English style—no one can deny its
charm or fail to wish there were even more of it. Nothing is so
picturesque and nothing so cheerful of aspect as the black and white
work which forms so frequent a feature in the earlier buildings
erected. One only regrets that it is difficult to justify it from a strictly
commercial point of view, especially if it is executed in a sound and
substantial manner. Whether the half-timber work is used for the
whole building, or only partially in connection with the fine red
sandstone of the district, or with bricks or flint-work, it has an
undeniable and enduring charm, and we owe much of our pleasure
in the whole appearance of Port Sunlight to the liberal views of the
founder, who did not permit his vision of a beautiful village to be
obscured by the clouds of philistinism! You cannot, of course,
pretend that such gables as those shown in our illustrations are
necessary to cottage building. Nor is it surely possible for even a
Port Sunlight to be entirely built in such a way; but the pleasure
produced by such character of work is, after all, common property,
and is a valuable item in regard to the whole scheme.

W. & S. OWEN, 21. A PICTURESQUE CORNER IN PARK ROAD SOUTH.

Architects.
 The Plan.
The general plan of Port Sunlight shows now an inhabited area
nearly a mile long by nearly half that wide, bounded on the longer
sides by the new Chester Road (on the east) and the main railway
lines to London, and Greendale Road (on the west). (See No. 39.)
There is enough variety of level to avoid the monotony of an entirely
flat area, and one piece of natural dell, well grown over with trees
and shrubs, forms a delightful feature near the Works end of the
village. Goods from the Works are loaded, on the one side, into
railway wagons, and on the other into barges on the Bromborough
Pool, from which they emerge into the River Mersey. From this pool
there used to be gutters or ravines, up which the muddy tidal water
flowed right up into where the village now stands, but these have all
been cut off from the tide and, with the exception of the dell above
referred to, filled up.
W. & S. OWEN, 22. BEBINGTON ROAD COTTAGES.
Architects.

One very notable innovation on the common practice of estate

development is the fronting of houses towards the railway instead of
the long lines of unlovely backs which usually exhibit all their
unhappy privacies to the railway passengers. Though one long
thoroughfare—the Greendale Road—runs alongside the railway
embankment for the greater part of a mile, one cannot feel it to be
other than one of the pleasantest roads on the estate. One of the
illustrations indicates the excellent result here obtained.
WILSON AND TALBOT, 23. COTTAGES, POOL BANK.
Architects.

Every intelligent student of town-planning knows that you cannot

rule out a number of rectangular plots arranged on axial lines
without due consideration of varying levels and a proper expression
of local features. Moreover, the planning of many right-angled plots
is not in itself a very desirable aim. But at Port Sunlight it was
possible to create some rectangular spaces with the Art Gallery and
the Church on their axial lines in such a way as to make a striking
and orderly scheme as a central feature in the estate. There are
numbers of winding or diagonal roads which give variety and interest
and afford pleasant lines of perspective to the groups of houses.
 DOUGLAS AND MINSHALL, 24. COTTAGES, POOL BANK.
Architects.

In an especial way one might claim that the best results in the
planning of a new village will be obtained through bearing in mind
the classical saying, “Ars est celare artem.” In such a scheme we do
not wish to be confronted with buildings of ponderous dignity or a
big display of formal lines and places. Anything approaching
ostentation or display is surely out of place, and what we want is
something expressing the simplicity and unobtrusiveness which is
the tradition handed down to us through the charm of the old
English village. This is best attained by variety in direction of roads
and shapes of houses by forming unexpected corners, recessed
spaces, and winding vistas.

W. & S. OWEN, 25. HULME HALL.

Architects.
 General Scheme.
Port Sunlight village (founded in 1888), apart from the Works,
covers 222 acres, on which the houses may approach 2,000 for a
population of 10,000. The tenancies of the houses are limited to
employés of the Works. Already over 1,000 houses have been built
or are in process of building, and the length of broad roadways
exceeds five miles. The first block of cottages built in 1888-1889 was
reproduced at the Brussels Exhibition of 1910, and was awarded the
Grand Prix. It is intended to limit the number of cottages to ten per
acre, and it is hoped to keep below that maximum.

W. & S. OWEN, 26. GLADSTONE HALL.

Architects.

The general width of the roadways is 40 feet, giving 24 feet to the

road, and 8 feet for each footpath; but there are roads 48 feet wide,
including footpaths. The paths are flagged along the central portion
only.
In a progressive world, and especially in such a progressive part of
it as Port Sunlight, one cannot hope to give a record which will for
long represent existing facts. The arrangements which have been
made for the benefit of the inhabitants of this village have
necessarily been altered or modified. At the present time the
buildings for general use include Christ Church (No. 6 and Pls. 31-
33), an admirable Late Gothic building in a central position, the
Schools, which accommodate about 1,600 children, a Lyceum, a
Cottage Hospital, a Gymnasium (No. 29), an open-air Swimming
Bath (Pl. 26), Post Office (Pl. 19), a Village Inn (No. 36 and Pl. 24),
Village Stores, a Fire Station, the Auditorium, to seat 3,000, the
Collegium (No. 11), the Gladstone Hall (No. 26), the Hulme Hall (No.
25), Co-Partners’ Club with billiard rooms and bowling green (No. 8),
a Village Fountain, and, finally, the Hulme Art Gallery (Pl. 29), which
is destined to hold the Public Library as well as fine collections of
Pictures, Pottery, and Furniture.
27. PARK ROAD BY POETS’ CORNER.

Port Sunlight has been an object of attraction to visitors for years,

and this is not only due to the interest and variety of its cottage
houses, and as a model for town planners the world over, but to the
whole-hearted endeavour to meet all the practical and social needs
of everyday life which is expressed in its various public buildings. But
another source of great and enduring attraction lies in its Art Gallery.
Here it outdistances every other village of the kind, for this Art
Gallery holds no fortuitous collection of odd things, but carefully
chosen examples of fine art got together by expert knowledge. The
pictures, china, furniture, etc., would alone bring many visitors to
study such a superb and finely-housed collection of works of art.
 Tree Planting.

DOUGLAS AND FORDHAM,

Architects.

28. BRIDGE COTTAGE.

We are apt to forget that a newly created village or town does not
reap all its benefits at once. Not only as regards the results of
growth in trees and shrubs, the development of gardens, and the
mellowing influences of time and tone, but also in relation to all the
amenities of social life, we must wait for those influences which can
only come in a gradual process. The subject of trees alone, of the
best method to deal with living growth, is not finished with for some
time, if ever. Some of the avenues at Port Sunlight are charming
now, and show an admirable balance of effect between trees and
buildings. Down the avenue of poplars one of our sketches (Pl. 3)
shows how delightfully the Club and the Library peep out, and how
well the vista leads up to the Post Office beyond—so in some of the
winding roads the effect even in summer is just right. But trees keep
growing, and unless the houses are to suffer they will have to be cut
down and some removed entirely. Then, again, the Diamond (Pl. 2)
(which in spite of its name is a great oblong open space), bordered
by groups of cottages and bounded at one end by the new Art
Gallery, will very well bear all the height the trees will ever reach.
This is a very fine open space, and borders of big trees will help, and
never belittle it. Possibly the secret of successful planting amongst
cottage houses is to have plenty of slow-growing evergreens, and
forest trees only at intervals. It is quite certain that if the garden
spaces at Port Sunlight were punctuated with decoratively placed
evergreens, and inclosed by living borders of box or yew, the result
would be both pleasing and long-lasting. The open spaces now
secured should make for ever pleasant oases amongst the long lines
of houses, and even if all the tree avenues had to go, there would
still be left much to excite the envy of those who have to live in our
dirty old towns.

W. & S. OWEN, Architects. 29. THE GYMNASIUM.

 One of our sketches shows the avenue which leads to Christ
Church from Greendale Road (Pl. 13). It is obvious that the sturdy
breadth and dignity of the church will never lose anything, however
lofty the avenue becomes. Unfortunately we cannot afford the space
in the thoroughfares for the trees so that they will not be a trouble
to the buildings some day. The only possible way would be to plant
them down the centre of the roads, so keeping the traffic in the two
opposite directions in its right place. This is a counsel of perfection,
but it has been done where wide road spaces were practicable.

J. J. TALBOT, Architect.
30. THE TECHNICAL INSTITUTE.

It will be noted that at either side of the Diamond the land round
and between the houses is bordered by a low wall through which
steps lead up to the pathways. The effect is very pleasing and might
be repeated in other cases with advantage.
 Cottage Plans.

31. WOOD STREET COTTAGES.

An evidence of the careful economic spirit which has guided the

whole enterprise may be found in the plans of the buildings at Port
Sunlight. There are here no freaks or features created simply for
picturesque effect, nor any serious attempt to give the occupants
something they do not want. It will probably be a long time before
any great reform in cottage planning can be maintained in face of
the varying views of the tenants. Thus the rooms must be big
enough, but they must not be so large as to cause needless work.
The better class cottages must have parlours, and only those who
cannot afford them will go without. Plaster walls seem to be almost
always preferred to those lined with boarding, white-washed bricks,
or any other healthy or artistic departure from the modern British
type. Thus we find that the compact and economic plans in the
village are what give the most universal satisfaction. But in the
scheme of the planning the juxtaposition of the cottages has been
dealt with in a free and varied manner, so that we find rows of
houses, or L-shaped blocks, or semi-quads, or curved frontages, or
semi-splayed quads. A census of opinion would probably be all in
favour of straight rows, and have been dead against the judicious
variety which gives so much interest to the place. Theoretically, one
would perhaps like those who live in cottages to give up the fetish of
the parlour and have one really ample living-room instead. But the
inherent yearning for privacy is an English characteristic which closes
the door of domestic affairs from the casual visitor. Moreover, the sin
of affectation creeps into all our buildings, and thus the cottage apes
the little villa, the little villa apes the large one, the large one apes
the mansion, and the mansion apes the palace.
 32. A GARDEN CORNER.

The cottage reformer would of course say that the cottage tenant
would be far happier and healthier as a rule without a parlour, for
then he would have a fine living-room which might be free of all
incumbrances and free of draughts. But it has to be taken for
granted that most who can afford parlours prefer to have them;
therefore the plans are of two types, the kitchen cottage and the
parlour cottage. Our illustrations show how these are planned, and it
is not of little interest to see how varied may be the exterior
treatment as developed from these plans.
 J. J. Talbot Architect

33. KITCHEN COTTAGES.

Some of the plans which have been found successful we give

illustrations of. These (Nos. 33-35) are carefully schemed. There is a
bath in each and three bedrooms, each with a fireplace. The W.C.’s
are entered from outside. The parlour cottage plan is also given. It
shows what a fine living-room might be obtained in a scheme which
eliminated the parlour. It is obvious that the question of cost is more
or less elusive. The original cost of the smaller cottages was £200,
and of the parlour cottages £330 to £350, but this has risen now to
£330 for cottages and £550 for parlour houses. At the present time
the gross rentals of the kitchen cottages average now 6s. 3d. each,
whilst for the parlour cottages the rent would be 7s. 6d., excluding
rates and taxes.[1]
 34. PARLOUR COTTAGES.

In any estimate of the value of Port Sunlight as a housing scheme

it must always be remembered, as Mr. W. L. George has pointed out,
that it is an experiment rather in ideal than in cheap housing. This
question of ideal was the first point referred to in this record. That it
has been largely realised and entirely justified is something for which
its founder must feel profoundly glad. All sorts of economies and
precautions might have been adopted which have been boldly and
generously set aside. The ideal was always kept in view, and if it
ever disappears it will be only after the disappearance of the original
founder himself! It is a pleasant task to gather together in this little
book the evidence of belief that a more real partnership between
capitalist and workpeople would work a lasting good. That good is
not to be measured in a notation of gold, nor even amongst those
who live and thrive under the immediate benefits of Port Sunlight.
Its influence goes round the world like the beneficent rays which are
symbolised in its own expressive title.

35. KITCHEN COTTAGES.

Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com