Automotive Safety Handbook PDF

Automotive Safety
Handbook
Second Edition
Ulrich Seiffert
and
Lothar Wech
Warrendale, Pa.
All rights reserved. No part of this publication may be reproduced, stored in a
retrieval system, or transmitted, in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written permission of
SAE.
For permission and licensing requests, contact:

SAE Permissions
400 Commonwealth Drive
Warrendale, PA 15096-0001 USA
E-mail: [email protected]
Tel: 724-772-4028
Fax: 724-772-4891
Library of Congress Cataloging-in-Publication Data
Seiffert, Ulrich.
Automotive safety handbook / Ulrich Seiffert and Lothar Wech. --
2nd ed.
p. cm.
ISBN 978-0-7680-1798-4
1. Automobiles--Safety appliances--Handbooks, manuals, etc.
2. Automobiles--Testing--Handbooks, manuals, etc. I. Wech, Lothar.
II. Title.
TL159.5.S45 2007
629.2’31--dc22 2007005180
SAE
400 Commonwealth Drive
Warrendale, PA 15096-0001 USA
E-mail: [email protected]
Tel: 877-606-7323 (USA/Canada)
724-776-4970 (outside USA)
Fax: 724-776-1615
Copyright © 2003, 2007 SAE International

SAE ISBN 978-0-7680-1798-4
SAE Order No. R-377
Printed in the United States of America.
Preface to the First Edition
This book on automotive safety describes the relevant development of safety

for passenger cars. Because of our long-standing experience in the field of
automotive safety (i.e., both authors have worked for more than two decades
in this field), basic relationships and new developments related to accident
avoidance and mitigation of injuries are described. Included in discussions
are driver support systems, chassis, lights, body and interior design, restraint
systems, biomechanics, dummies, accident simulation tests, pedestrian pro-
tection, and compatibility.
This book is suitable for those who are interested in safety engineering, and
for students and experts who are interested not only in details but also in the
broad perspective of vehicle safety.
Special thanks are extended to the many persons and companies supporting
this book with relevant material, especially Autoliv, Bosch, Continental,
DaimlerChrysler, and Volkswagen.
vii
Preface to the Second Edition
The second edition of this book is an update based on changes in the legisla-
tion and technology, especially in the areas of accident avoidance, pedestrian
protection, and compatibility.
Table of Contents
Table of Contents
1. Introduction ...................................................................................... 1
1.1 Reference ..................................................................................... 1
'H¿QLWLRQV ......................................................................................... 3

2.1 References ................................................................................... 6
3. Driving Forces for Increased Vehicle Safety ................................. 7

3.1 References ................................................................................... 9
4. Safety Legislation ............................................................................11

4.1 Reference ................................................................................... 16
5. Accident Data ................................................................................. 17

5.1 References ................................................................................. 23
6. Accident Avoidance ........................................................................ 25

6.1 Human Factors........................................................................... 25
6.2 Comfort and Ergonomics........................................................... 29
6.3 Acceleration and Braking .......................................................... 43
6.3.1 Adaptive Cruise Control (ACC) ...................................... 47
6.3.2 Brakes .............................................................................. 48
6.3.3 Brake-by-Wire ................................................................. 52
6.3.4 Vehicle Dynamics ............................................................ 56
6.4 Information Systems.................................................................. 63
6.5 References ................................................................................. 66
7. Biomechanics and Occupant Simulation ..................................... 71

'H¿QLWLRQ ................................................................................... 71
7.2 Injury Tolerance Limits ............................................................. 73
7.3 External Injuries ........................................................................ 74
7.4 Internal Injuries.......................................................................... 76
7.4.1 Concussion....................................................................... 76
7.4.2 Spinal Injuries.................................................................. 79
7.4.3 Chest Injuries ................................................................... 80
ix
Automotive Safety Handbook
7.5 Criteria in the Rule-Making Process ......................................... 82

7.5.1 Head Protection ............................................................... 82
7.5.2 Chest Protection............................................................... 83
7.5.3 Neck Injury ...................................................................... 84
7.5.4 Performance Criteria for the Rule-Making Process......... 86
7.5.4.1 Chest ................................................................... 86
7.5.4.2 Pelvic .................................................................. 87
7.5.4.3 Leg and Knee...................................................... 87
7.5.4.4 Pedestrian Protection .......................................... 89
7.6 Test Devices............................................................................... 89
7.6.1 Body Part Test Devices.................................................... 91
7.6.1.1 Head Impact.........................................................91
7.6.1.2 Torso Impact ........................................................91
7.6.1.3 Pedestrian Accident Simulation Tests..................94
7.6.2 Three-Dimensional Dummies.......................................... 94
7.6.2.1 Frontal/Rear Collision and Rollover....................97
7.6.2.2 Lateral Impact......................................................97
7.6.2.3 Child Dummies....................................................99
7.6.2.4 Rear Impact Dummy (Bio-RID)........................100
%LR¿GHOLW\'XPP\ ............................................102
7.6.3 Human and Dummy Modeling ...................................... 102
7.7 References ............................................................................... 106
8. Vehicle Body ................................................................................. 109

8.1 General..................................................................................... 109
8.2 Low-Speed Impact....................................................................117
8.3 Vehicle Body Test Without the Simulation
of Car Occupants ......................................................................119
8.3.1 Quasi-Static Test Requirement .......................................119
8.3.1.1 Seat and Seat-Belt Anchorage Point Tests........ 119
8.3.1.2 Roof Strength.................................................... 121
8.3.1.3 Vehicle Side Structure ...................................... 122
8.4 References ............................................................................... 125
9. Dynamic Vehicle Simulation Tests .............................................. 127

9.1 Frontal Collisions .................................................................... 127
9.1.1 Pole Test......................................................................... 132
9.1.2 Frontal Car-to-Car Crash ............................................... 133
x
Table of Contents
9.1.3 Design Requirements of Frontal Collisions................... 136

9.2 Lateral Collisions..................................................................... 138
9.3 Rear-End Collisions................................................................. 145
9.4 Rollover ................................................................................... 145
9.5 References ............................................................................... 149
10. Occupant Protection .................................................................... 151

10.1 Vehicle Compartment .............................................................. 151
10.2 Restraint Systems .................................................................... 151
10.2.1 Seat Belts...................................................................... 152
10.2.2 Airbags......................................................................... 160
10.2.2.1 Airbags for Frontal Impacts .......................... 160
10.2.2.2 Side Protection by Airbags............................ 169
10.2.2.3 Additional Airbag Applications..................... 170
10.2.2.4 Sensors for Restraint Systems....................... 172
10.3 Child Restraints ....................................................................... 179
10.4 References ............................................................................... 182
11. Interrelationships Among Occupant,

Restraint System, and Vehicle in Accidents ............................... 187
11.1 Frontal Impacts ........................................................................ 187
11.1.1 The Unrestrained Occupant.......................................... 187
11.1.2 The Three-Point Belt.................................................... 189
11.1.2.1 The Seat-Belt Clamping Device Above
the Retractor ...................................................192
11.1.2.2 The Mechanical Pretensioner .........................192
11.1.2.3 The Pyrotechnic Pretensioner.........................192
11.1.2.4 The Seat-Belt Load Limiter............................194
11.1.3 Passive Restraints......................................................... 196
11.1.4 The Airbag System....................................................... 198
11.1.5 Comparison of Test Results for Different
Restraint Systems..........................................................200
11.2 Lateral Collisions..................................................................... 200
11.2.1 Theoretical Analysis for Lateral Collisions.................. 202
11.2.2 Side Airbag Systems .................................................... 206
11.2.2.1 Thorax Side Airbags ..................................... 206
11.2.2.2 Side Head Protection Airbags....................... 210
11.3 Rear-End Collisions................................................................. 212
xi
11.4 Rollover Protection.................................................................. 216

11.5 Special Requirements and Opportunities for
Sensor Applications ................................................................. 218
11.6 General Literature.................................................................... 221
11.7 References ............................................................................... 221
12. Pedestrian Protection .................................................................. 225

12.1 General..................................................................................... 225
12.2 European NCAP-Test .............................................................. 231
12.3 Legislation Activities............................................................... 231
12.4 Technical Solutions for Vehicles in Accidents......................... 240
12.5 References ............................................................................... 243
13. Compatibility ................................................................................ 245

13.1 Theoretical Analysis ................................................................ 251
13.2 Research Analysis Proposals, Consumer Information,
and Legal Requirements .......................................................... 256
13.3 Conclusions ............................................................................. 262
13.4 References ............................................................................... 263
14. Computer Support for the Development of Safety

Components .................................................................................. 267
14.1 The Basics................................................................................ 267
14.2 Description of the Numeric Tools............................................ 268
14.3 Calculation of Components ..................................................... 269
14.4 Total Vehicle Crash Computation............................................ 273
14.5 Occupant and Restraint System Simulation ............................ 275
14.6 Pedestrian Simulation Tests..................................................... 278
14.7 Summary.................................................................................. 278
14.8 References ............................................................................... 278
Index ...................................................................................................... 281
About the Authors ................................................................................ 293
xii
Introduction
Today, "safety and security" in all our activities has become more and more
a basic element in our day-to-day experiences. There are several reasons for
this development. The largest influence is the increasing number of people
in the world. As a consequence, more resources are being used for daily liv-
ing than the Earth might be able to provide. Other parameters are the rapid
rate of change in technolgies (e.g., modem communication techniques), the
use of all kinds of energy, increasing social differences, the consumption of
fossil fuels, local and global accidents, and crime. This list could easily be
made much longer. Because information systems are available 24 hours per
day worldwide, knowledge about these factors and dramatic events is much
greater today than in the past.
For this reason, it is not surprising that in marketing questionnaires related

to customer wishes for new cars, when they are asked which performance
characteristic is most important in their decision-making processes for buying
a vehicle, safety is on the top of the list. This means that customers desire not
only safe and comfortable driving without breakdowns, but also integrated
information and communication systems and to be as safe as possible in the
event of an accident. Figure 1.1 shows that vehicle safety is one ofthe important
areas in customers' overall requirements for cars [I- I].
This book describes the most important areas of vehicle safety in passenger
cars. To work in the field of automotive safety is one of the most positive
activities to which an engineer can commit during his or her career.
1 . 1 Reference
1- 1. Seiffert, U. Fahrzeugsicherheit VDI-Verlag, Diisseldorf, 1992, ISBN 3-
18-401264-6.
n Safety
Figure I . 1 Conflicting demands by vehicle buyers.

Definitions
There are many definitions for vehicle safety. One of the best is based on
the statement made in 1955 by the German university professor, Professor
KoeDler: "The motor-vehicle has the duty to transport humans and goods
from place Ato place B, as safely, as quickly, and as comfortably as possible."
In modern life, we would add to this definition "...and as efficiently and
environmentally friendly as possible." Figure 2.1 illustrates a more scientific
perspective of vehicle safety [2- 11.
Vehicle safety can be subdivided into three main areas:
1. Accident avoidance
2. Mitigation of injuries
3. Pre-crash measures
Some additional definitions related to vehicle safety are as follows:
Accident avoidance. (Colloquial = Active safety.) All measures that

serve to prevent accidents.
Mitigation of injuries. (Colloquial = Passive safety.) All measures that

help to reduce injuries during accidents.
Pre-crash. All measures to reduce the severity of an accident after the

sensor detects that a severe accident is unavoidable.
External safety. Design of the external parts of the vehicle to reduce

injuries in the event of a collision with an external collision partner.
Interior safety. Design of the vehicle interior parts to prevent additional

injuries in the event of contact with an occupant of the vehicle.
Automotive Safety
Mitigation of injuries
vrj-rn
i
I
Human Vehicle Environment
physical, psychologica)
Traffic routing Protection of Protection Protection of

Acceleration own occupants
occupants and pedestrian
r-l
Traffic density Occupant protection
Accoustics
Traffic control
characteristics (e.g., dashboard, during frontal, Folding mirrors
steering wheel) lateral, rear-end
I Information systems1
Figure 2.1 Automotive safety
Definitions
Restraint systems. Vehicle components (e.g., seat belts, airbags) that

specifically influence the relative movement of occupants in relation to
the vehicle.
Smart restraints, sensors, and actuators. These apply to occupant

detection and pre-crash evaluation.
Primary collision. Collision of the vehicle with another obstacle.
Secondary collision. Collision of the occupant with vehicle parts.
Active devices. Safety features that must be activated manually for use
(e.g., active seat belts).
Passive devices. Safety and restraint systems that in an accident are acti-
vated automatically (e.g., airbags, pretensioners in seat belts, automatic
height-adjusted headrests).
The two main areas of safety-accident avoidance and mitigation of inju-

ries-often are called active safety and passive safety, respectively, in the
public domain. Although these phrases described the situation clearly until
the end of the 1970s, they are no longer valid today. Many passive safety
systems, such as airbags and seat-belt pretensioners, can react in a very
active manner, and the two areas are moving closer together. For example,
near-distance radar sensors could be used for several purposes, including the
inflation of airbags and the activation of seat-belt pretensioners.
In the field of accident avoidance, three main influences are found:
1. The human being

2. The technical features of a car
3. The environment
In many studies, human beings are cited as the main cause of accidents. If
this statement is examined in more detail, you will find strong influences from
other areas (e.g., comfort, noise, information technology [cellular phones],
and the man-machine interface) that might change this high number. In any
case, the driver is influenced by his or her mental and physical conditions, as
well as any drugs and/or alcohol that he or she may have consumed. With
respect to the vehicle, the general handling characteristics, driver support
systems (e.g., antilock brake system [ABS], automatic cruise control [ACC],
electronic stability program [ESP], power steering, automatic transmission,
and information systems), field of view, lighting, and comfort level (e.g.,
noise and vibration conditions, heating and air conditioning levels) influence
the behavior of the vehicle [2-21. To some extent, the road (i.e., layout, road
surface, and traffic signals) and the weather also play important roles. Some
functions in the field of accident avoidance are being combined with items
from the field of mitigation of injuries. One example is sensor fusion, where
sensors are used for both accident avoidance and as the pre-crash signal for
airbags (i.e., obstacle detection). Information systems can help to reduce the
occurrence of accidents and can notify an ambulance and the police automati-
cally if a severe accident occurs.
The field of mitigation of injuries can be subdivided into two main groups:
1. Protection during the crash

2. Protection after the accident
These two groups can be further subdivided into the protection of the vehicle
occupants in both vehicles in a multi-vehicle collision and, in a single-vehicle
crash, the protection of the occupants of that vehicle. Another important field
is post-crash safety, with several requirements such as the capability to open
the doors of the vehicle without tools and the prevention of fire.
A new area is the pre-crash situation. If the sensor system identifies that a
severe accident will occur, several activities are performed (e.g., automated
braking, pretensioning of the seat belts, the sunroof closed, the seat back in
a more upright position).
2.1 References
2-1. Seiffert, U. "Future Research Directions for Enhancing Safety," 19th
ESV Conference, Washington, DC, United States, 2005.
2-2. Braess, H.-H. and Seiffert, U. Handbuch Kraftfahrzeugtechnik, Vieweg-

Verlag, Wiesbaden, Germany, 2005.
Driving Forces for Increased
Vehicle Safety
Many factors and activities have helped to reduce the number of accidents on
all types of roads. The following are examples, not listed by priority:
Customer demand
Science
Public demand
Technology
Government legislation
Product liability
Consumer information
Competition among car manufacturers
Automotive press reports
Legislation certainly has a strong influence on safety. In Germany, with the

development of the first cars, a proposal for a law about liability for the use of
vehicles was introduced in 1909. In 1832, England introduced requirements
for a steam-powered bus with respect to performance during accidents. A
strong positive effect of legislation surfaced in rule-making activities in the
United States during the middle of the 1960s and during the International
Experimental Safety Vehicle Conference, where safety became a competi-
tive item for medical and technical scientists, legislators, and automotive
engineers.
The vision of government activities related to safety is very strong (basis

2005):
The Netherlands and Sweden: Zero fatality vision

United States: 33% reduction in fatalities by 2008
Canada and the European Union: 30% reduction in fatalities by 2010

United Kingdom: 40% reduction in fatalities by 2010
Japan: 50% reduction in fatalities by 2015
A second effect was the activity initiated by consumer advocates such as

Ralph Nader in the United States, and the increasing number of consumer
information reports with respect to the performance of vehicles in the field and
in accident simulation tests, such as the crashworthiness rating report issued
by the Insurance Institute for Highway Safety (IIHS) [3-11, the Highway
Loss Data Institute [3-21, and the worldwide New Car Assessment Program
(NCAP) crash tests [3-31. Table 3.1 shows an overview of the various NCAP
activities in Europe, the United States, Japan, and Australia, together with
those of IIHS.
TABLE 3.1
WORLDWIDE NCAP TEST [3-51
Euro NCAP U S . NCAP llHS A NCAP J NCAP
Rigid wall full - 56 kmlh - 56 kmlh 55 kmlh
frontal i m ~ a c t HIII HIII HIII HIII HIII HIII
Offset 64 kmlh - 64 kmlh 64 kmlh 64 kmlh
deformable HIII HIII HIII HIII HIII HIII HIII
barrier (EEVC) P3 P 1 112 P3 PI112
frontal impact
Mobile barrier 50 kmlh 62 kmlh - 50 kmlh 50 kmlh
side impact EEVC barrier crab barrier EEVC barrier EEVC barrier
Euro-SID II SID Euro-SID II Euro-SID II
PI112 P3 SID PI112 P3
Side pole 29 km/h - - - -
impact flying floor
Euro-SID II
Pedestrian 40 km/h - - - -
bodyform adult head
impacts child head
upper leg
lower leg
The information available to the public about the performance in the preceding
tests and about other criteria is provided not only by government organizations
[3-41 but by automotive magazines [3-51-even by the German auto enthusi-
asts' magazine, Auto, Motor, and Sport. Interested consumers also can find
Driving F O I ~forS Incl~easedVehicle Safety
the results on various websites. Because of consumer interest for information

in addition to the data that are already available, more information related to
vehicle performance with respect to accident avoidance is under discussion
for distribution. In addition to the two parameters already mentioned, a third
may be the most important one. This is the increasing interest by customers,
who rank safety features when buying new cars as "extremely" or "very"
important. When customers were asked about the importance of car safety,
the percentage of people who showed high interest in vehicle safety increased
from 64% in 1981 to 84% in 1999 in the United States [3-61. In Germany,
companies such as Audi and Volkswagen have published figures that state that
85% of people who were asked mentioned safety as the top issue. In the early
1990s, safety became a competitive item among automobile manufacturers.
This is one of the best things that could happen for both the consumer and
the field of safety. The increasing interest of all groups mentioned may be
connected with the longing for a secure and safe life.
Although not discussed in public intensively, product liability legislation also

has had some important effects related to the design of cars. In product liability
cases, the state-of-the-art defense leads to the fact that safety features, which
are introduced by one manufacturer, create pressure on other manufacturers
to install these new safety devices as standard equipment on all cars. The
high level of attention to road safety also has increased the number and the
quality of accident analyses performed by different institutions in various
countries. This is true not only for government organizations but also for
insurance companies, scientific and medical organizations, and some of the
major car companies.
References
Crashworthiness rating reports, various issues, sporadically published
by the Insurance Institute for Highway Safety, Arlington, VA, United
States.
Highway Loss Data Institute, www.carsafety.org.
Klanner, W. "Status Report and Future Development of the Euro NCAP

Program," Proceedings 17 Experimental Safety Vehicles (ESV) Confer-
ence, 2001.
3-4. Crashworthiness rating reports, various issues, sporadically published

by the National Highway Traffic Safety Administration (NHTSA) and
the Federal Motor Vehicle Safety Standards (FMVSS), Washington,
DC, United States.
3-5. "Safety Steps in the Spotlight," Automotive News, March 6, 2000.
3-6. Insurance Institute for Highway Safety, Special Issue: "Vehicle Compat-
ibility in Crashes," October 1999, and several publications in 2006.
Safety Legislation
The different government legislative institutions that are responsible for

the improvement of traffic safety are important because they combine three
elements:
1. Traffic routing (e.g., road design and traffic signals)

2. The education of traffic participants (e.g., drivers, school children)
3. Vehicle performance
The most important requirements do not describe the design but the perfor-
mance criteria that must be fulfilled in defined tests. Only such rules support
the creativity of engineers and the competition of different ideas.
With respect to automotive legislation on a worldwide basis, there are differ-

ent starting points for various countries. For example, the meaning set forth
by the responsible rule-making bodies in the United States was that traffic
participants, especially the driver, could be educated to only a limited extent;
thus, vehicle occupants must be protected in the event of an accident caused
by the vehicle itself (design).
The second priority was given to measures for accident avoidance. Contrary
to the U.S. perspective, European legislators gave much more responsibility
to the driver because the focal point was the prevention of accidents. In the
beginning and until the mid-1960s, the number of accidents with severe and
fatal injuries reached such a high level that the American government defined
in November 1966, with the Motor Vehicle Safety Act [4-11, intensive require-
ments for automobiles. Europe and other countries followed. Meanwhile,
several of the 100 requirements were introduced worldwide (Figure 4.1).
1. ECE
2. EG
3. Japan
4. Sweden
5. Australia
6. Germany
[*New Edition]
Year
Figure 4.1 Number of rules for vehicles in various countries.

(Source: VolkswagenA G.)
In Europe, the situation with respect to the rule-making process is as follows.

Vehicles that are able to drive faster than 6 krnlh (3.7 mph) and would be
driven on public roads must fulfill national requirements or, if possible, apply
for a European Economic Community (EEC)-type approval. If the vehicle
manufacturer applies for an EEC-type approval, it also must prove that it has
introduced a quality control system to assure the conformity of production
(COP). A typical application field is Germany. Table 4.1 shows the rules
for cars with up to eight seats for Europe EEC, Germany (StVZO), and the
United Nations Economic Commission for Europe (ECE) for accident avoid-
ance and the minimization of injuries. Table 4.2 shows the relevant rules in
the United States [4- 11.
Based on the agreement of 1958 with respect to common rules for the approval
of parts and vehicles, the United Nations Economic Commission for Europe is
working to harmonize the different rules worldwide. With the adjustment of
Safety Legislation
TABLE 4.1
REQUIREMENTS FOR ACCIDENT AVOIDANCE IN EUROPE
EC ECE
Directive Regulation StVZO*
Requirements for Active Vehicle Safety (Accident Prevention):
Steering equipment
Brake systems
Replacement brake padslshoes
Equipment for acoustic signals
Field of vision
Defrosting and defogging systems for glazing
Windshield wipers and washers
Rearview mirrors
Heaters (engine waste heat)
Gas heaters, auxiliary heaters
Installation of lighting and lighting-
signaling devices
Reflex reflectors
Clearance lamps, taillamps, stop lamps
Side marker lamps
Turn signal lamps
Headlamps for high beam andlor low beam
and their light sources

Gaseous discharge headlamps
and their light sources
Front fog lamps
Rear fog light (fog taillamp)
Backup lamps (reversing lamps)
Parking lamps
Rear registration plate illumination devices
Reverse gear and speedometer equipment
Interior equipment (symbols, warning lights)
Wheel covers
Tire tread depth
Tires and tire mounting
Towing capacity, hitch vertical load
Towing equipment (trailer hitches)
Pedal arrangement
"(StraOenverkehrs-Zulassungsordnung = Road Traffic Licensing Regulation)

TABLE 4.1 (cont,)

Requirements for Passive Vehicle Safety (Mitigation of Injuries):
Interior fittings (protruding elements)
Steering mechanism (behavior in an impact)
Frontal impact, occupant protection
Side impact, occupant protection
Seat-belt anchorages
Seat belts and restraint systems
Seats, seat anchorages, head restraints
Head restraints
External protection
Fuel tanks and rear underride protection
Liquefied petroleum fuel systems
Doors (locks and hinges)
Front and rear bumpers
Rear-end collisions (not applicable to
Germany)
Child restraint systems
Safety glazing
Electric propulsion (safety)
TABLE 4.2
RELEVANT RULES IN THE UNITED STATES [4.1]
FMVSS Contents
Controls and displays
Transmission shift lever sequence, starter interlock, and
transmission braking effect
Windshield defrosting and defogging systems
Windshield wiping and washing systems
Hydraulic and electric brake systems
Brake hoses
Lamps, reflective devices, and associated equipment
New pneumatic tires
Tire selection and rims
Rearview mirrors
Safety Legislation
TABLE 4.2 (cont.)

FMVSS Contents
Hood latch system
Theft protection
Motor vehicle brake fluids
Power-operated window, partition, and roof panel systems
New pneumatic tires for vehicles other than passenger cars
Tire selection and rims for motor vehicles other than passenger
cars
Air brake systems
Accelerator control systems
New non-pneumatic tires for passenger cars
Passenger car brake system
Occupant protection in interior impact
Head restraints
Impact protection for the driver from the steering control system
Steering control rearward displacement
Glazing materials
Door locks and door retention components
Seating systems
Occupant crash protection
Seat-belt assemblies
Seat-belt assembly anchorages
Windshield mounting
Child restraint systems
Side impact protection
Roof crush resistance
Windshield zone intrusion
Fuel system integrity
Flammability of interior materials
Fuel system integrity of compressed natural gas vehicles
Compressed natural gas fuel container integrity
October 16, 1995, the working title now has the following meaning: "New
Agreement concerning the adoption of uniform technical prescriptions for
wheeled vehicles, equipment, and parts that are used in road vehicles ..."
Application for membership is on a voluntary basis. This means that countries
that are members of the United Nations but not of the EEC could join this
organization.
Since March 1998 and November 1999, the EEC and Japan, respectively,
became members of this commission. More than 100 rules are in effect, and
some of those rules could be used as part of the EEC-type approval. Addi-
tional actions to minimize trade barriers among the continents of the United
States, Japan, and Europe started with the Transatlantic Economy Dialog.
The goal is to improve trade relations between Europe and the United States
and to reduce trade barriers. A further commitment was given to create, on
a more global basis, rules for vehicles. In the International Harmonized
Research Activities (IHRA), the following themes, led by the countries
below, are handled:
Australia: Side impacts

Italy: Frontal impacts
United Kingdom: Vehicle compatibility
United States: Biomechanics
Canada: Intelligent transportation systems
Japan: Pedestrian safety
Guided by the IHRA Steering Committee, the number of harmonized rules

should be increased. As of February 2005, approximately 80% of the rules
were capable ofbeing harmonized, if the car manufacturer accepts the higher
level of safety in the other continents.
4.1 Reference
4- 1. Compiled from sources of the National Highway Traffic SafetyAdmin-
istration (NHTSA) and the Federal Motor Vehicle Safety Standards
(FMVSS), Washington, DC, United States.
Accident Data
This chapter provides some general data related to accident statistics. In addi-
tion to this general data, we can say that for more complex engineering work
on vehicle safety, very detailed information is necessary. Worldwide, many
different organizations are actively involved in accident research work, such
as governments, universities, insurance associations, and vehicle manufac-
turers. In direct meetings among the experts and in many conferences, the
facts and conclusions are discussed.
Figure 5.1 shows the development of traffic fatalities as a hnction of billion

vehicle kilometers for various countries from 1970 and 1980 up to 2003.
Although the rate of improvement is impressive, too many accidents continue

to occur, with severe and fatal injuries on public roads. On a worldwide basis,
there are more than 1 million fatalities per year on roads. In Europe and the
United States, one could observe approximately 45,000 fatalities in 2005. If
we look more closely at the distribution of fatal accidents, Figure 5.2 shows
the various participants in Germany as a function of years. The group of
passenger car occupants is the largest, although it also shows the highest
degree of improvement.
Another strong influence is the road itself. Figure 5.3 demonstrates the dif-
ferences in the types of roads in Germany.
This information is relevant in absolute numbers and if you compare the

fatalities per kilometers driven where, for example, the safety per kilome-
ters driven is approximately a factor of four better on highways than on
other roads. There is one special effect in Germany and in other European
countries for country roads. Approximately 30% of all fatalities on coun-
try roads occurred because of a large concentration of impacts with trees.
How this number will be reduced as a result of the installation of electronic
I -1
l i m 1
Japan U
United States u
France U
United Kingdom
Germany
/'
-0
0 5 10 15 20 25 30 35
Fatalities per billion vehicle kilometers
Figure 5. I Development of the,fatality rate in road trafic on highways

in Japan, the United States, and Europe (fatalities per billion
vehicle kilometers). (~ource:R$ 5- 1.)-
pedestrians
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Year 19..120..
Figure 5.2 Nzrmber of deaths in accordance with trajic participants in

Germany. (Source: Ret 5-2.)
Accident Data
8,000
Germany West Germany afier reunification incl. former East
Source: StBa
German Federal Autobahn
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004
Year 19..120..
Figure 5.3 Number of deaths as a,function ofroad type.

(Source: Ref 5-2.)
stabilization control and head side airbags will be seen in the future, but a
positive effect can already be observed.
Figure 5.4 shows an international comparison with respect to the reduction

of traffic deaths per billions of kilometers driven on highways in various
countries.
There also is no clear indication that countries with speed limits on highways
have a lower accident rate than others, in terms of fatal accidents.
If we examine the type of crash for car occupants, we also find interesting
results, as shown in Figure 5.5 [5-2, 5-31. In the United States, 40% of all
crashes are single-car accidents. This also is true for Germany. However,
we find differences in areas such as cars to pickup trucks and sport utility
vehicles, pedestrians, and two-wheeled vehicle accidents. It also is important
to look at the driver, who often is the main cause of the accident. In addition
i.2
Jap 4.0
Ser 3.2
Fra 2.9
1970 1975 1980 1985 1990 1995 2000 2005

Year
Figure 5.4 Death rate on highways-an international comparison.

(Source: Re$ 5- I .)
Accident Data
(4
Distribution o f P a s s e n g e r V e h i c l e A c c i d e n t s in G e r m a n y i n 2 0 0 3
Pedestrian &
Bicycle
- I -
Truck
Passenger Vehicle
1
-
a
m
C
Motor~zedTwo-wheeler -
I
0
Single-Vehlcle Accident
Figure 5.5(a) Distribution of accident types in the United States.

(b) Distribution of accident types in Germany. (Source: Re$ 5-2.)
to the more direct parameter, which is related to the specific items discussed
in Chapter 6 on accident avoidance, a general observation could be made:
a correlation definitely exists between age and risk. Figure 5.6 shows the
results of a study conducted in the United States.
From Figure 5.6, we can conclude that young drivers have a high risk for caus-
ing crashes. This high risk will be increased by the number of passengers in the
car [5-41. A study conducted in Germany, as shown in Figure 5.7, compares
the number of pedestrians and car occupants injured in accidents.
0 1 2 3 + 0 1 2 3 + 0 1 2 3 +
Ages 16-17 Ages 16-19 Ages 3G59
Number of Passengers
Figure 5.6 Crash rates by driver age and passenger presence.

(Soz~rce:Rex 5-4.)
Figure 5.7 shows a high risk in both categories, that is, accidents per kilo-
meters driven and accidents per time for children and young people [l-I].
Another question, which often is asked in private discussions, is whether
women or men have more or fewer accidents. As illustrated in Figure 5.8,
there seems to be a general advantage for women with regard to a lower
number of accidents, especially in single-car crashes.
Accident Data
- C P w s e n g e r t a r Cibef
. - - - - -- - -
34
+Pedestrian
- C F a S s e n g C r Car OIIver
2
0
0.5 6.9 10.74 15.:7 18.70 71,24 15-94 45.61 65 74 174
Years
Age g m w
Figure 5.7 Accident rates for pedestrians and passenger car drivers in
Germany (Source: Rep I - I.)
References
International Road Traffic and Accident Database, Bundesanstalt fiir
Strapenwesen, Bergisch Gladbach, Germany, 2005.
Verkehr in Zahlen 2004, Deutscher Verkehrsverlag, Bundesministerium

fur Verkehr (Ed.), Hamburg, Germany, 2005, ISBN 3-87154-314-4.
Insurance Institute for Highway Safety annual report, Vol. 34, Febru-
ary 1999, Arlington, VA, United States.
Auto Motor Sport, 2 1/ 1999, Stuttgart, Germany.

'a . Car-Car Accidenls (Men)
- -
500 - b Car-Car Acc~denls(Women)
Single-Car Accidents (Men)
-Single-Car Accidenls (Women)
18-M 21-24 2534 35-44 4554 5554 > 54 years
Age In Years
Figure 5.8 Passenger car accidents per 1,000 owners of drivers licenses
and one million kilometers driven. (Source: R e r 5-5.)
Accident Avoidance
In the category of accident avoidance fall all measures that help in a positive
manner to prevent accidents. Related to the vehicle driver, we can state that also
the contribution for an easier, safer, and more comfortable drive helps to avoid
accidents. Driver assistance systems should support the driving function.
In addition to the technical features of drive-by-wire, it is very important to the

man-machine interface to define precisely the responsibilities that the driver
should retain. The responsibility of the driver also is one basic requirement
of the Wiener world agreement for "Driver Assistance Systems." It is easy
to understand that this does not have to be a rigid definition. Technological
progress and a growing understanding of the man-machine interface will
allow more continuous step-by-step progress in this area. One fundamental
rule must be fulfilled: New technical features should not be installed in any
case if those features have a foreseeable negative effect, even if this effect
is very small.
6.1 Human Factors

Humans as traffic participants play a major role with respect to failure that
might lead to an accident. Table 6.1 shows the involvement of various groups
related to the occurrence of accidents.
The main group is car occupants. Some data indicate that the driver is respon-
sible for more than 80% of all accidents. If we take an average of all German
states, we find the data in the police reports for 200412005 with respect to the
group that is responsible for these accidents, as shown in Table 6.2 [6-11.
Although the driver's group is the most significant as the major cause of
accidents, we must be careful not to neglect other influences. The driver
TABLE 6.1
DISTRIBUTION OF TRAFFIC PARTICIPANT GROUPS
WlTH RESPECT TO ACCIDENTS
Type of Europe Germany United States
Traffic Participant ["/.I ["/.I w.1
Passenger car occupant 42-63 51 73.5
Motorized two-wheeler 14-27 14 7.5
Bicycle driver 5-20 9 2.0
Pedestrian 14-34 23 14.5
TABLE 6.2
THE ORIGINATOR OF ACCIDENTS WlTH INJURIES,
AS A FUNCTION OF TRAFFIC PARTICIPANTS
Responsible for the Accident %
Driver (excluding cyclists) 77.0
Cyclist 9.1
Pedestrian 5.6
Road conditions 5.1
Technical failures on cars 1.O
Others 2.0
is influenced by his or her health and physical conditions, qualifications and

experience, driving education, and ability for orientation; quality of the
man-machine interface; climate; and comfort related to the vehicle. In more
detailed studies, we also find different figures for the driver being responsible
for causing the accident. For example, a vehicle with accident prevention
measures or higher comfort levels also might lead to a lower accident rate. If
we take this into consideration as a result of the analysis of various sources,
we find approximately 65% human-based failures, 30% environmental, and
5% vehicle-related. [f we examine a detailed analysis of the single failures
for the driver, the general distribution for the 2003 statistic in Germany [6-11
could be taken from Table 6.3.
Accident Avoidance
TABLE 6.3
CAUSES OF ACCIDENTS FOR ALL TYPES OF VEHICLES,
INCLUDING MOTORCYCLES, BIKES, AND OTHER VEHICLES
Driver-related:
Speed too high
Right of way (intersections)
Turns, drive in and out, turn around
Driving too close to another car
Influence of alcohol
Misuse of the road
Failure during passing
Wrong behavior against pedestrians
Others
Vehicle itself
Pedestrians
Road conditions
Others
The mental and physical condition of a driver is one of the most important
factors to consider when discussing the avoidance of accidents. Mistakes
made by the driver have varied causes. Often, he or she dismisses the pos-
sibility of an accident, a momentary over-estimation of self, lack of driving
experience, insufficient physical condition, and impairment due to the con-
sumption of alcohol or drugs. Alcohol consumption affects the safe driving
ability of the driver.
Figure 6.1 shows an investigation with various people and blood alcohol
content (BAC), driving a specific course on the Volkswagen driving simu-
lator. The test persons had to drive a certain distance, where the number
of mistakes as a function of BAC was measured. A significant deviation
from normal driving was observed with an alcohol content of approxi-
mately 0.6 0/00. In a study of accidents, the increase in the risk to cause
an accident as a function of BAC is as shown in Table 6.4.
Driving error F = exceedina drivina lane (sum over total time) x 100%
driving time
Figzwe 6.1 Driving failur*es as a,firnction o f blood alcohol content (BAC).

(Source: Ref 1-1.)
TABLE 6.4
RISK INDEX OF ALL ACCIDENTS AS A FUNCTION
OF BLOOD ALCOHOL CONTENT (BAC) [6-21
BAC Level Risk Index K
0 1.o
Accident Avoidance
In most countries, we find legislation limits on blood alcohol content as

defined by the authorities. From the first work on experimental safety
vehicles, there have been many investigations on preventing the drunk driver
from being able to start the vehicle. All invented devices until now have
failed. Either these devices were so complicated that even drivers without any
blood alcohol content were unable to start the car, or drunk drivers overcame
the devices with the help of a "solver" person. In addition to this negative
influence of alcohol, we find drugs as a cause of accidents, although the exact
figures are unknown.
Another element of research is that tired drivers may fall asleep while driv-
ing the vehicle. Drowsiness sensors might be one solution for these critical
situations. For example, systems would watch the movements of the eyelid
via cameras and give an acoustic or mechanical (vibration) warning signal.
Other possibilities would be to monitor the steering wheel movement or the
pulse frequency and breathing of the driver.
In general, we can state that for hture work on vehicle safety, detailed knowl-
edge about the behavior of the driver and the other traffic participants from the
results of a more detailed accident investigation becomes increasingly important.
However, for new systems, only a theoretical prediction of the effect can be
made, and extensive studies on driving simulators must be performed.
Some single elements of accident avoidance measures are described in the

following sections.
6.2 Comfort and Ergonomics

To a certain degree, there is a direct relationship between comfort and vehicle
safety. All aspects that contribute to a more comfortable drive also help to
reduce accidents. This includes comfortable entry and egress from the vehicle,
which has the entrance and seat not too low above the road, a good adjustable
seat, and a seat back with lumbar support. Several other elements, which
vehicle occupants today take for granted, contribute to an increased level of
comfort. These include seat-belt latches on the seat, height adjustments of
the upper anchorage point, and vertically and horizontally adjustable steering
wheels. Figure 6.2 illustrates a connection between the comfort of the seat
and driver fitness.
Brain aupplled with adequate amount
Visual power sufficient
ood hearing ability
ood flexibility of the neck

Heart and blood circulation intact;
blood pressure In an appropriate range
Jolnts movable
healthy; diabetes handled with
medlcatkm
- I
No congestion in the leg veins
1\
Upper leg support not too rigid;
Body weight proportionallydistributedto
p l v i s and upper legs
relaxed muscles
Figwe 6.2 Accident avoidance measzves. (Source: Re6 I - I .)
Other minor items can have a great positive influence on driver comfort.
These include electrically adjustable and heated mirrors, power windows,
power steering, heated wiper systems, rain sensors, and automatically adjust-
able interior rearview mirrors to avoid blinding from glare. Although he1
consumption due to the higher amount of electrical devices in the car often
increases slightly, the benefit for safety justifies these installations. One posi-
tive example is the much higher installation rate of air conditioning systems
in cars. For many years, this was the domain of cars in the United States,
Japan, and the upper-class segment in Europe. Meanwhile, in many coun-
tries, air conditioning in cars has become standard equipment. As shown in
Figure 6.3, from an investigation performed by the company Behr [6-31, we
can find a direct correlation between the accident rate and the heat loading
on the driver.
The weather likewise directly influences the occurrence of accidents. On wet

roads, we find a 20% increase in accidents, with a high thermal load slightly
above a 20% increase.
Another comfort item is low interior noise. Figure 6.4 shows the dB (A) level
of an optimization process due to the installation of insulation measures in a
vehicle body as a function of engine revolutions per minute (rpm).
Accident Avoidance
% change in average number of accidents
Wet Road Moderate Heat Stress High Heat Stress
Increase in number of in-town accidents in case of wet roads and heat stress of the driver.
Figure 6.3 InJEuence o f weather conditions and heat loading of the driver:
t 000 2000 3000 4000 5000 6000

Speed (llmm)
Figure 6.4 Reduction o f the total noise pressure level due to

inszrlation measures. (Source: Ref: 6-4.)
In this connection, controlled measures for noise and vibration reduction are
becoming necessary to eliminate special problems. Not only is a low level
of noise important, but also the fact that single spikes should be avoided. In
addition to a low level of interior noise, an optimized climate control in the pas-
senger compartment, with the possibility of guiding the airflow to individual
parts of the human body (and on high-class cars, the ability of the driver and
passenger to adjust the climate control separately) is standard equipment.
The vertical oscillations of the vehicle also influence the comfort level of
the driver and the vehicle occupants. Figure 6.5 demonstrates the relation-
ship between oscillation and damping comfort and contact between tire and
road. The optimal layout is within the limited values (e.g., below a K-factor
of 10).
K-Value,, (mmfort scale)
Fdyn,, I Fstat (F= wheel load)
Figwe 6.5 K-scale as function o f wheel loud and vehicle

natural frequency. (Source: R e t 6-5.)
According to Richter [6-51, a good compromise for the layout of the chas-
sis is shown, if the data as shown are fulfilled, although the necessary low
unsprung mass could not always be achieved as easily. Engineering research
and development activities today are leaning toward the development of the
Accident Avoidance
electrlc and air sp1n9and
block
sensor
Figure 6.6 System architecture for an electronic suspension system.

(Sozvce: Ref: 6-6.)
active chassis, which also reduces the critical oscillation frequencies. Figure 6.6
shows the design of an active chassis control.
The ergonomic design for the working-place "driver" and for the man-machine
interface is evident with regard to accident avoidance capability. In the BMW
700 series, the redesign of the operating concept tried to integrate the follow-
ing features of the intuitive and adaptive man-machine interface [6-71:
Intuitivism
Transparency
Failure robustness
Effectiveness
Individuality
Safety
Learning ability
Flexibility
Adaptive
Socially compatible
Multi-modularity
The reaction of the market today favors the new solution.
The working-place driver also must be designed for different sizes of humans.
The two-dimensional design dummies cover the 5% female up to the 95%
male, as shown in Figure 6.7.
95% Male
7 5% Female
Figure 6.7 Steering wheel andpedal layout in a microbus.

(Source: Ref: 1-1.)
From studies at the University of Kiel (Germany), we know that an aver-

age man of the young generation (i.e., 20 to 25 years old) has a total body
length that is approximately 12 cm (4.7 in.) larger compared with the group
of people between the ages of 16 to 60. For the actual layout of the interior
of the vehicle, car manufacturers meanwhile are using more realistic data.
Because of a large number of legal requirements, a three-dimensional test
device is used, as shown in Figure 6.8 [6-81.
Accident Avoidance
Figure 6.8 Description of the three-dimensional machine tools.
With this device, the seating reference point can be determined. The seating
reference point is the basis for the field of view, for the seat-belt anchorage
point location, and for other vehicle parameters. 1n addition, designers use
various mathematical models to simulate vehicle occupants. Computer tech-
nology today allows a much better simulation of the ergonomic design of the
working-place driver. In the virtual product creation process, different sizes

and three-dimensional effects can be simulated. Figure 6.9 demonstrates
RAMSIS, a three-dimensional computer model simulating the driver [6-81.
Figure 6.9 Instrzrment panel layout b-Y RAMSIS.
For the layout of the interior of the vehicle, not only are the driver's physical
dimensions considered, but the interior lighting systems are receiving more
attention. During the entry and egress process of vehicle occupants, lighting
should contribute to safety and orientation. During driving, the driver should
not be disturbed by the reading light or the instrument panel lights. The instru-
ment panel should be read easily by the driver and have a color that is optimal
for most drivers. The instrumentation should not produce reflected glare in the
windshield. Although head-up displays give the designer an additional free
parameter, these have not achieved a large market penetration. This might
change if more warning functions are given via head-up displays to the driver.
Initial solutions are already on the market.
Accident Avoidance
In terms of the driver's field of view, numerous legal requirements exist.

The driver should have a good all-around view, related to the vehicle itself,
and a free field of view through the windshield. The requirements in the
field of view and the wiped pattern on the windshield are described, for
example, in FMVSS 103 and 104 [6-9,6-101. The field of view is divided
into Fields A, B, and C. Figure 6.10 shows these areas for a compact car.
Figzwe 6.10 Thefield of view for a compact passenger car:
The areas in the field of view are defined by tangential surfaces on the ellipse
of the driver's eyeballs. The defined areas must be wiped with a specified
percentage (more than 80% for Field A, more than 94% for Field B, and
more than 99% for Field C). Similar requirements must be fulfilled in other
countries such as Europe. The determined areas also serve for the checking of
the performance of the windshield heating system (defrosterldefogger). The
vehicle under investigation is tested in accordance with test procedure SAE
5902 [6-111. The vehicle is pre-conditioned at an environment temperature

of -1 8°C (-0.4"F). In a period of less than 40 min, Field A must be defrosted
by more than 80% and Field C by 100%.
The all-around view includes the exterior and interior mirrors. 1n the area
of optimizing the field of view, we also find new wiper systems that by their
geometric layout minimize the unwiped area. Several support systems also
are available in some cars. These include heated wiper nozzles and heated
outside mirrors, electrical mirror adjustment, electrically heated windshields,
rain sensors (see Figure 6.1 1) that automatically switch on the windshield
wipers, and automated interior anti-glare rearview mirrors.
( I ) LED
(2) Pholodlode
(3) Optical F~ber
(4) OpUcal Coupling
(5) Drop
(6) Ambienl Lghr Sensor
(7) Wndshleld
Figure 6.11 Layozrt o f a vain sensor: (Sozrrce: Ref 6-12.)
Amore advanced system uses sensors that are installed in the outside mirrors to
observe both sides of the car, to prevent an accident in a critical lane-changing
maneuver, or to prevent a possible accident with a pedestrian or cyclist in a
right-turn situation. It has not yet been determined how the critical situation
should be transmitted to the driver. Several solutions are possible, ranging
from an acoustical warning signal to an increase of steering wheel torque, as
shown in Figure 6.12.
Other systems that indirectly improve the field of view are obstacle iden-
tification systems that measure at low-speed driving the distance to other
obstacles and provide an acoustical or light signal to the driver, as shown
in Figure 6.13. The following functions could be fulfilled [6-131:
Accident Avoidance
CCD camera
Figure 6.12 Rearview milProrwith integrated camera electronics,for
detecting objects in vehicle blind spots. (Soz~rce:Re$ 6-13.)
rear
Figure 6.13 Microwave sensor concept for vehicles. Sensor,

video senso6 and laser scanner (Sotwce: ReJ: 6-13.)
View of blind spot

Parking assistant
Adaptive cruise control (ACCtShort-range sensor for stop-and-go function
Pre-crash detection
The connection to vehicle safety is given by the fact that small amounts of
damage to vehicle exterior components and therefore to the lights and signals
are prevented.
One example of a large number of research activities to improve the driver's

all-around view is demonstrated in Figure 6.14 by the Bosch company. If
we analyze Figure 6.14 closely, we could imagine that the large number of
sensors are able to improve the safety of the vehicle in several areas if the
man-machine interface is designed correctly. The driver assistant systems can
support several functions, such as the parking pilot, with ultrasonic sensors
in the bumper system, up to 1.5 m (5 ft); adaptive cruise control (ACC); a
long-range radar sensor of 77 GHz; and a short-range sensor of 24 GHz for
identifying obstacles close to the vehicle. In the future, cameras will be used
Protection zones
77 GHz Infrared Video 24 GHz Ultrasound

Long-Range Radar Short-Range
(Lidar) Radar (Lidar)
Remote distance Extended Medium range Short range Ullra short range
middle distance 0 bis 5 80 m
1 m b e -120 m 0 bis r, 200 rn 0.2 b ~ s .20 m 0.2 bis .: 1.5 (2.5) m
Figure 6.14 Vehicle s~u-~~otlnd

sensom (Source: ReJ: 6-14.)
Accident Avoidance
in addition to sensors. This provides an opportunity not only to measure the

distance to objects in front of the vehicle but to identify of the types of objects.
Other sensors might provide information about road and bad weather condi-
tions such as snow and fog.
In relation to the area of field of view, we also must add the function of the
signals and the lights of vehicles. The signals are not used to improve vision
but to identify the vehicle and to show the purpose of the driver's action, if
he or she changes the direction of the drive path or applies the brakes. For
passenger cars, a third brake light at an increased height above the ground is
installed. This light allows drivers in the vehicle behind the braking vehicle
to more quickly recognize a braking situation, thereby helping to speed up
their reaction time to apply the brakes. There are also some discussions
regarding whether the third brake light should change its color intensity in
relation to braking force.
With respect to the development of headlamps, Figure 6.15 shows the changes
as a function of years. Newer and powerful LED developments already allow
rear lamps and in the future will allow headlamps for cars.
Related to the design of headlamps, the trend to install the technical elements
behind a glass or plastic cover is continuing. The shape of the headlamp cover
is designed for low aerodynamic resistance and pedestrian protection.
For the layout of headlamps in terms of lighting intensity, we must make sure
that the headlamps provide an optimized field of view for the driver, without
blinding other drivers on the road. In general, we can ask for the fulfillment
of the following requirements:
The road should be illuminated symmetrically and without spots.
The driver should have a good view of the road shoulder on his side.
The majority of light should cover the driving path.
Scatter width should be sufficient.
Avoid blinding the oncoming traffic.

Figure 6.15 Product development of'headlamps. (Source: Refi 6-15.)
Accident Avoidance
The driving path in front of the vehicle up to a distance of 40 m (13 1 ft)

should not be illuminated to an extreme level because of the possibility
of blinding from a wet road, due to the reflection of light from oncoming
traffic.
Also in the lighting equipment, we find some driver assistance systems. For
example, the automatic vertical adjustment of headlamps is especially requested
for xenon light. The automatic headlamp leveling device must be able to
identify the position of the vehicle relative to the road. This is accomplished
by the use of angular sensors. In the future, even more advanced systems
are foreseeable. One example is additional infrared headlamps to identify
pedestrians or other road obstacles in darkness.
In addition to vertical adjustment, a computer-controlled horizontal headlamp

adjustment based on either the steering wheel angle or, in the future, in com-
bination with a positioning sensor such as GPS (global positioning satellite)
is possible. The benefit of this solution lies in the fact that road curves are
illuminated before the vehicle has reached the next curve of the road. Every
year, more production cars with curve lighting systems are entering the mar-
ket. The adaptive front lighting will have several dynamic functions. One
function is the town light, which illuminates a larger area compared to the
asymmetric beam used today. The other function is the country light, which
will include the recognition of road course obstacles and road signs in a more
efficient manner, as shown in Figure 6.16 [6-161.
6.3 Acceleration and Bralung

Because of the big difference among passenger cars with respect to engine
power and vehicle mass, and because of the necessity to keep traffic flowing,
establishment of minimum performance with respect to power and torque is
necessary. This is a request not only for Europe but for North America and
some states in Asia. For most passenger cars, front-wheel drive is standard.
To achieve sufficient traction and brake capability, many additional features
are available in production cars. For traction control, we find that different
names for the system are used by the car producer. In principle, this system
uses a portion of an antilock brake system (ABS) for electronic differential
control. This device is sufficient to prevent a traction wheel from spinning
when starting a vehicle on a surface with a different coefficient of friction.
Conventional AFS
Figure 6.1 6 Advancedfrontlight system (AFS) (dynamic curve light).

(Source: Ref: 6- 16.)
At higher speed (e.g., above 20 kmlh [12 rnph]), the effect is reduced. Above
40 kmlh (25 mph), the system is switched off to avoid a negative influence
with respect to safe driving under high speed. Figure 6.17 shows the general
function of such a system.
With other systems (e.g., automatic stability control plus traction [ASC+T]),
during a wheel spin of one drive wheel, the propulsion force via an electronic
engine management system is reduced. In addition, through a brake inter-
vention at the spinning wheel up to 40 kmlh (25 mph), this effect is again
reinforced. The electronic differential system (EDS), propulsion slip control
(ASR), and ASC+T systems are sufficient for many driving conditions. A
more sophisticated device to control the acceleration capability is four-wheel
drive. Several four-w heel-drive systems are on the market. Figure 6.18 is an
overview of the different driveline systems, including four-wheel drive.
Four-wheel drive shows positive function not only under acceleration and
complicated road conditions but during aquaplaning (hydroplaning) as well.
Accident Avoidance
Figure 6.1 7 An antilock brake system/propt~lsionslip control system (ABS/

ASR [acceleration slip regulationjl,for passenger cars. ( I ) Speed sensor;
(2) ABS/ASR hydrazdic unit; (3) ABS/ASR ECU; (4) Electronic lhrottle
ECU; (5) Throttle gasoline engine. (Source: Rej 6-1 7.)
FA Front-wheel drive
D-V-D Permanent all-wheel drive (slip controlled)

Viscose-clutch with freewheel
D-D-D Permanent all-wheel drive (power controlled

with differential gear) MJM, = 111
D-DV-D Permanent all-wheel drive (power and slip

controlled) MJM, = 112
D-S-D Rigid all-wheel drive
D-S-S Rigid all-wheel drive with locked rear differential
SA Standard drive
Figure 6.I8 Eew of driveline systems. (Source: Ref.' 1-1.)
45
A pioneer for permanent four-wheel drive for passenger cars was Audi AG.
Volkswagen's system is based on a hydraulic electronically controlled clutch,
where the front and rear axle are connected, if this is required by unusual road
situations, as shown in Figure 6.19.
3 Multi-plates
r
-----
4 Axial-piston pumps
, 1 5 Control valve
1
1 O u t ~ ushafl
l
Figure 6.19 Fzrnction of the four-wheel-drive Haldex clutch.

(Source: Rej.' 6- 18.)
For four-wheel drive cars, which are used primarily on normal roads, the four-
wheel system is disconnected to avoid vehicle spinning if, during downhill
driving, the brakes are applied.
For most owners, the vehicles equipped with EDS/ASR systems are sufficient
in day-by-day driving. This is the reason why four-wheel drive remains
a niche market, although high-powered vehicles and sport utility vehicles
(SUVs) have a higher and increasing installation rate of four-wheel drivelines
in the vehicles.
Accident Avoidance
6.3.1 Adaptive Cruise Control (ACC)

Adaptive cruise control (ACC) is one system that is already in series produc-
tion. It started in connection with upper-class models and now is available
in the compact class as well. Adaptive cruise control identifies the vehicle
in front of the ACC-equipped vehicle, determines its position and speed, and
maintains (through braking and acceleration) the desired distance between
the two vehicles. The main benefit is achieved on roads with constant high
speeds, such as highways and the Autobahn in Germany. To maintain the
desired speed and safe distance from the vehicle driving in front, the car
equipped with ACC uses a 77-GHz radar sensor to identify moving objects
to a distance of 100 m (322 ft) in front of the vehicle and at an angle of + 4 O
to the centerline. With the determined data, the electronic control unit cal-
culates the relative distance to the moving object in front of the ACC-equipped
vehicle. If no other vehicle is in the front of the ACC-equipped vehicle, the
vehicle accelerates to the selected speed. Although the driver receives strong
support for this drive function, he or she remains responsible for control of the
vehicle. Figure 6.20 shows some parts of the ACC by Bosch (the radar and the
electronic control unit). In the newest PRE-SAFE activities, DaimlerChrysler
has started to offer a more advanced system called the PRE-SAFE brake.
Technical Data
Temperature range -40 to 80°C
Power consumption 13 W
Package size: 12.4 cm H
9.1 cm W
9.7 cm D
Weight 10.6 kg
Bracket for vehicle-specific
mounting including alignment
mechanics
Figure 6.20 Components of adaptive cruise control (ACC).

(Source: Ref: 6-20.)
Around 2.6 sec before an accident, the driver receives an acoustic warning
signal; around 1.6 sec before an accident, the vehicle automatically applies
the brakes, with a deceleration of up to 4 m/sec2 [6-201. Other companies
are installing similar systems.
6.3.2 Brakes
Standard equipment on passenger cars today is a hydraulic two-circuit brake
system with two independent working brake circuits. A typical layout is the
diagonal configuration of the brake (one for a front and a rear wheel). Chas-
sis design with a negative kingpin offset supports directional stability during
the braking maneuver. A load- or pressure-dependent braking force regula-
tor prevents locking of the brakes at the rear axle. This guarantees stability
during braking.
Figure 6.2 1 shows the importance of a short reaction time and a rapid increase
in brake pressure for the prevention of accidents.
Some production cars use a so-called "brake assistant." In emergency braking

situations, due to a pressure boost control or via a dual rate booster with an
emergency valve, this brake assistant rapidly increases brake pressure, even
Collision Speed
Figure 6.21 Distribution of accidents. as aji~nctionofcollision speed

(Source: Rec 6-2 I .)
Accident Avoidance
at relatively low brake-pedal force. The design of the brake assistant can be
described as follows:
The signal of the pressure sensor determines from the pressure level and pres-
sure gradient if an emergency braking situation exists. This increases the braking
force, if necessary, in an initial phase. If the pressure applied by the driver is
reduced, the brake pressure is increased automatically again. The brake pres-
sure also is increased if the driver is not pushing the brake pedal hard enough.
Figure 6.22 shows details of the control function [6-221. The brake assistant in
a not-too-aggressive layout is being introduced in more vehicles.
As already mentioned, the PRE-SAFE system together with ACC auto-

matically applies the brakes if an accident is predicted. Audi with its SUV-Q7
[6-231 uses the brake pedal movement to force the driver to activate the brakes
immediately to prevent an accident with the vehicle in front. The sensors used
are already installed in the vehicle by the ACC system and give to the driver first
an acoustic and visual warning. lf an immediate brake application is necessary,
a momentary brake bounce requests the driver to apply the brakes.
Performance of the brakes to decelerate the car and simultaneously maintain

the steering capability is a very important factor. The FMVSS 105 and EEC
Directive 711320 describe many design and performance criteria. Each
vehicle manufacturer also has its own additional test programs such as down-
hill performance during a long descent from a high mountain and extensive
winter testing.
Basic requirements are two-circuit brakes, a brake booster, a braking force

regulator for the rear brakes, sufficient brake cooling, and multiple brakes
out of high speed without a significant reduction ofbrake performance. Also,
the perfect function of the brake system during the lifetime of the vehicle is a
very important parameter for safe driving. As mentioned, the self-stabilizing
effect at the front axle is one important element to keep the vehicle in line
during braking on roads with different coefficients of friction from side to
side of the vehicle. Figure 6.23 shows a cross section of the wheel contact
area (midpoint), the spring strut, and the track rod.
The view also shows the stability function. In a braking situation at the wheel
with the higher brake force, a moment arises that corrects the tendency to
Deceleration
......................
W~thbfb
....... ....................
\ '
.......................
i Slow driver
........................... react~on
.........................:.....................................
................................................. .....................................
.....................t . . . . . . . . . . .:.. ..............
........ ."........................................ ...............
Time Is]
Force at brake force booster (bfb)
Input Force
Figure 6.22 Brake assistant. (Source: Rej.' 6-22.)

Accident Avoidance
-11- Stabilizing kingpin offset
Figure 6.23 Cutaway view through the wheel-road

contact area. (Source: Re$ I - I .)
steer in the direction of this wheel; thus, the vehicle track remains stable.
For vehicles, it also is important that the rear wheels do not lock during
braking. This is achieved, when necessary, by a brake pressure reducer or
with the ABS. State-of-the-art for the ABS is the three-channel system.
Figure 6.24 shows the schematic function. Via electromagnetic valves, the
brake pressure is controlled at the single wheel to avoid the locking of one
of the specific wheels.
The opinion often mentioned by the public is that ABS reduces, in any case,
the braking distance. However, this is not true. The main purpose of ABS
is the capability to steer and brake. The span of control of ABS allows a
ABS control cycle for high adhesion

coefficients.
islip switching threshold 2 . 1
----+
Time t
Figtrre 6.24 Function o f an ABS conti-ol (one wheel).

(Source: Reb 6-24.)
variation of 8 to 35% slip between a single wheel and the road surface. With
this data, the brake distance is optimally short, but there is enough side force,
which is necessary for steering.
6.3.3 Brake-by-Wire
The preceding examples have shown the importance of electronic brake man-
agement related to accident avoidance. Two systems are being investigated
Accident Avoidance
to use brake-by-wire technology for brake systems. The first application in

a production car is the sensotronic brake control (SBC), an electro-hydraulic
brake system (EHB) [6-251. Figure 6.25 shows the general layout.
sure supply
ngle piston pump
ressure accumulator
Pressure controller
I * Wheel individual
Front Rear
Closed loop control
- axle axle
Figure 6.25 Sensotronic brake control (SBC). (Source: Re$ 6-25.)
The systems contain the following components:
Activation unit, including the brake pedal
Hydraulic control unit with a specific electronic control unit, which could
be installed in a location separate from the activation unit
Yaw rate sensor, including a lateral acceleration sensor
Rpm sensors at the wheels
Steering wheel angle sensor

With this system, greater design freedom for the brake functions is possible.
The result shows shorter stopping distances, reduced pedal effort, and a bet-
ter feeling at the brake pedal (e.g., no vibration of the brake pedal as occurs
with conventional ABS). Also, optimization of ABS, ASR, ESP (electronic
stabilization program), and ACC is much easier. Likewise, the system is more
convenient through smart stop. It also will be one of the important features if
automated driving on roads becomes available in the future. In an accident,
it is advantageous that the tandem master cylinder used in the system has a
shorter length in front of the brake pedal. Therefore, any intrusion into the
passenger compartment could be minimized. Figure 6.26 shows the optimized
brake functions by SBC. The deceleration versus pedal movement shows
that braking must be a compromise between too "poisonous" and too "blunt"
performance. If a system failure occurs, a fall-back situation is given through
the fact that with the remaining hydraulic circuit, the two front wheels can
be used to brake the vehicle.
With the SBC (EHB) system, several additional functions are possible. For
example, in critical situations where emergency braking is necessary, the
Electronic brake pedal

electronic control of the whole braking system
1 Electrically driven power supply
I 0 High-pressure accumulator
Controllable ball valves
* Closed loop control of hydraulic pressure

in the wheel brakes - wheel individual
-
Hydraulic backup
1 lowest degraded mode of the braking system
Figure 6.26 Optimized brake function by SBC. (Souire: Ref 6-25.)

Accident Avoidance
brake assistant increases the brake line pressure until the antilock braking
controller reacts. Also, a pre-filling of the brake lines is possible in the case
where the acceleration pedal is relieved with a jerking motion in anticipation
of the need for braking. Even as the rain sensor senses the start of a rainfall,
the braking system with unnoticeable brake impulses can remove a water or
salt coating from the tires. After installation in some vehicles in Europe, the
system was not able to obtain market penetration, mainly because drivers
could not identify improved performance.
The second advanced system is the electronic mechanical brake system

(EMB). Figure 6.27 shows the layout in principle.
The basic difference with the EHB is that the hydraulic unit is replaced by an
electric motor with a mechanical adjustment unit at each wheel to apply the
brakes via an electrical signal. It is evident that the possible design freedom of
Central bearing
Brake caliper Motor
\
solver
\
",
Brake disc Spindle
Figure 6.27 Czltaway view o f an electronic mechanical brake system

(EMB). (Soz~rce:Rec 6-26.)
this system is even greater. On the other hand, the control unit must be fail
safe; that is, without electric power, the system should not be able to lock the
brakes. In addition to this requirement, we need at least a second independent
electric circuit, including the battery, to ensure that the brakes continue to
perform if one electrical circuit does not work. Some interesting pre-devel-
opment work is underway at the Siemens Automotive division, where work
is being done on the electrical wedge brake [6-271.
6.3.4 Vehicle Dynamics

The front-wheel-drive system has the highest market share compared to other
vehicle layouts such as standard rear-wheel drive and four-wheel drive. The
front-wheel-drive system offers the driver comfortable steering and a nicely
tuned chassis, even in small vehicles. Also in this area, we find, in addition to
hydraulic power steering, electrical devices to decrease the torque at the steer-
ing wheel and to reduce fuel consumption. For a good straight-ahead drive,
both the front and rear axles must be designed to hlfill the requirements for
a good-natured, forgiving but precise and comfortable chassis. Figure 6.28
shows the front and rear axles of a modern compact passenger car [6-281.
At high speeds, it is necessary to minimize vertical lift at the rear axle. In

the design of the vehicle, we must find a good compromise between low
aerodynamic resistance and low rear-axle vertical lift. Figure 6.29 shows
the aerodynamic coefficient, CD, versus the rear axle lift coefficient, CAH.
If the rear-end lift is too high, especially during lane-changing maneuvers at
high speed, this might become critical. The correlation between spring and
damping comfort and sufficient contact between tire and road also is important.
This correlation was shown here previously in Figure 6.5.
Although we have improved theoretical knowledge and better simulation tools,

the knowledge of experienced test engineers is still used for chassis design and
layout. Numerous tests, as shown in Table 6.5, are used to determine vehicle
behavior under dynamic situations [6-291.
One special test in Europe was invented after much public discussion about
the behavior of vehicles in extreme driving maneuvers. Although the ESP
(Electronic Stabilisation Programme) was already under development, the tests
and the public discussion about a so-called "moose (elk) avoidance maneuver"
Accident Avoidance
Figtwe 6.28 Front (top) and rear (bottom) axles of a compact cau:
(Source: Ref 6-29.)
Rear-end lift coefficient CLR
0.28 0.30 0.32 0.34 0.36 0.38 0.40 0.42 0.44

Drag coefficient CD
Figure 6.29 Rear-end I$' coeficient, as a function of drag coeficient.

(Source: Re$ 6-4.)
definitely accelerated the installation of the ESP system in production cars.

The ESP supplements the following systems already installed:
Antilock brake system (ABS)

Electronic brake booster (EBV)
Electronic differential system (EDS)
Antiskid control device (ASR)
The yaw moment controller uses the measured variables: wheel velocity,
yaw velocity, steering wheel angle, lateral acceleration, and pre-pressure at
the brake master cylinder. With the input of the steering wheel angle and the
vehicle velocity, a setpoint value is determined and compared to the signal
of the yaw velocity and lateral acceleration sensor. If any differences exist,
the yaw moment control unit generates signals that control the brake pres-
sure of each individual wheel, as well as the engine and transmission ECU.
The resulting forces at the single wheels allow stabilization of a vehicle that
has a tendency to skid. The ESP is the first system that supports the driver
to a large extent, compared with systems that can be judged to be activated
by the driver. One design-known as ESP-Plus-uses three wheels to
control the vehicle function. Figure 6.30 demonstrates the principle. After
Accident Avoidance
TABLE 6.5
EXAMPLE OF SINGLE CRITERIA FOR CHASSIS CONTROL
Subjective Assessment of Driving Behavior
1. Drive-Away Behavior 4.Cornering Behavior
I .1 Squat 4.1 Cornering behavior
1.2 Drive-away oscillation 4.2 Turn-ln abillty
1.3 Drive-away shaking 4.3 Lateral force increase
14 Steerlng input 44 Yaw velocity increase
1.4.1 Coefllcient of friction. High 4.5 Transverse control ability
1.4.2 Coefllcient of friction: Split 4.6 Roll behavior
1.5 Torque-sleer 4.7 Diagonal dip
1.6 Sleerlng jam 4.8 Suppoll effect
1.7 Traclion 4.9 ROII screening
1.7.1 Coefllcient of friction: High 4.10 Lane-change performance
1 7.2 Coefllclent of friction: Low 4.11 Steer-brake pertormance
1.7.3 Coefllcient of friction: Split 4.12 Steer-acceleration performance
1 7.4 Coefllcient of friction: Sudden change 4.13 Road impacl
1.8 Control response ATC 4.14 Load alteration effect
1.9 Pedal rear travel ATC 5. Straight-Running Stability
2. Braking Performance 5.1 Straight ahead
2.1 Braking Decelerat~on Spring steering
Coefllcient of Friction: High Roll Steering
Coefllclent of friction: Low Steer oscillation
Coefllcient of friction: Split Ridging
Coefltcient of friction: Sudden change Track rut sensitivity
Slabiliiy Load alteration sleerlng effecl
Slraight ahead stabilily Side wind sensitivity
Cornering stability Wmd sensitiv~ty
Steer stability Trailer wobblino
Yaw slability 6.Driving Comfort
Coefllcient of friction: High 6.1.1 Ride comfoll. low soeed
Coefllclent of Frotlon: Low 6.1.2 Ride comfoll. high speed
Coefllcient of Iriction: Splil 6.2 Pitch behavior
Brake dive 63 Roll behavior
Pedai elforl 6.4 Body damping
Pedai feel~ng 6.5 Rolling comforl
Pedal reaction ABS 6.6 Harshness
Pedal movlng ABS 6.7.1 Rolling noise
Brake ludder 6.7.2 T ~ r ewhining
Brake noise generai 6.8 Edge sensitivity
Squeal 6.9 Roar
2.14 Tramp 6.10.1 Thump~ng
3. Steering Behavior 6.10.2 Damper
31 Pivoting 6 11 Bouncing
3.2 Responsiveness 6.12 Absorb capability (bumps)
3.3 Trench effect 6.13 Rebound
3.4 Center point 6.14 Buffer stao
3.5 Sleering effort 6.15 Return
35I Central position 6 16 Tlrelfreeway hop
3.5.2 Proportional range 6.17 Springing
3 5.3 Parking 6.18 Stutler (5-15 Hz)
3.6 Sleermg passing 6.19 Load alteration
3.7 Overshoot 6.20 Body vibration
3.8 Post-oscillalion 6.21 Steering vibration
3.9 Poststeerlng 6.22 Steering shimmy
3.10 Targel precision 6.23 Sleer~ngbounc~ng
3.11 Road contacl 6.24 Steerlng return kick
3 12 Maneuverabilily 6.25 Steerlng ratlle
3.13 Steerlng return 6.26.1 Sealing comfort isolation
6 26 2 Seal lateral supporl
Figure 6.30 Principle o f layoz~to f an ESP system.

(Source: Rex 6-33.)
several years in production, the ESP shows very good performance in rela-
tion to accident avoidance and has reduced the number of lateral collisions
significantly [(i-301.
Another system that offers a compromise between safety and vehicle dynamics
is active chassis control, which also allows a high degree of driver comfort.
For example, the active chassis could influence in a positive way the roll
gradient of the vehicle [6-3 11.
Other factors are important in the design of accident avoidance means, such
as the tire and rim. Figure 6.3 1 shows an advanced system that even includes
a sidewall torsion sensor [6-321.
In addition to the specific performance requirements for tire and rim design,
tire pressure is a key element with respect to fuel economy and vehicle safety.
Because most customers do not carefully monitor tire pressure as frequently
as they should, the tire pressure will be reported to the driver by sensors
installed at each wheel. Different technical solutions are under development.
For example, if the tire pressure becomes too low, radio signals transmit the
signals to the instruments in the dashboard. One example is shown by the
Beru RDKS, which also offers an aftermarket solution. This system consists
of wheel electronics, valves, sensors, antennas, and an ECU [6-341. Tire
pressure will be monitored in the future in an increasing number of vehicles
on the road.
Accident Avoidance
Development of Modern Tires: The Tire as an Integrated Element

Advanced Tire Technology; Interaction BrakeTTireIRoad
Frlction model
P = 11 (v, T, PI slip) '.,
-. Road surface
Tlre models:
descrlptlon of texture
, ABS ( d r y h t ) 1 blocked 1 (fractal contact model)
,
..----,
lnteractlve
-.
1
system
-
tire brake
tlre-vehlcle
(slmulatlon) @ -
(design,
Brakecontrol
system ...)
- -
Combination of different R8D areas

such as chem~stry,mechancs
Sidewall s~mulat~on sensoncs, electronics.
torsion sensor wnlrolllnq, and brake technolow lo
come to a new interacttve brake
system.
Figzwe 6.31 Sidewall torsion sensor: (Sowce: Rer 6-32.)
Traffic-guidance and
information systems
i
Traffii lights Ad
I 1 Lane markings
1 Road Tri
Coord. Trattit
Figure 6.32 Driver assistance and trafic guidance systems: a chronology.

(Source: Re$ 6-36.)
61
It is evident that all electronic and drive-by-wire systems must be more precise
in their development with respect to software-and hardware-security, as
described, for example, in Ref. 6-35. In the field of accident avoidance, many
innovations are already in production vehicles available today, as shown in
Figure 6.32 [6-361.
The future will hold even more systems such as active lights, active steering
systems, active gas pedal, and extended information systems such as floating
car data and car-to-car and car-to-infrastructure communication. As men-
tioned, these systems will be successful only if the man-machine interface is
designed in such a way that the driver is integrated into the appropriate loop
while retaining control of the vehicle. This was demonstrated in a positive
way, in a research vehicle by Delphi [6-371, although actual production is
years away.
The integrated safety system (ISS) defines increased safety in a broader

approach and combines collision avoidance and mitigation of injury systems
with electronics. The key to future vehicle systems in this regard is the suc-
cessful integration of the following technologies:
Positive solutions for the man-machine interface
Functional multimedia systems, such as telematics
Driver workload management (attention)
360" sensors for collision warninglintervention systems (e.g., lane change,

blind spot, and vehicle control), adaptive cruise control (ACC), andlor
lateral guidance
Control of the chassis component (comfort and accident avoidance)
Advanced safety interiors (smart restraints)
X-by-wire technology
Accident Avoidance
I Lateral Guidance
- 'Actwe -7 ~ r o n t l , i g h t ~ ~?n -! 7
-F?W
System (AFS) AL~iomatic Driving'
Ernelgehcy Braking Electronic,
ESP Distance Control (ADR) I?brake-by-Wire -. -
Go-Pilot
Collision Avoidance
--
Reauirements
Sensor Technology
Signal Processing
System Safety
System Integration
Test and Qualification
Figure 6.33 Evolution o f a road map o f driver assistance systems.

(Source: Ref 6-13.)
After the installation of ABS, ESP, curve light, and lane-keeping devices,
brake activation in case of a possible accident without any action by the driver
might have a high priority in passenger cars, trucks, and buses.
6.4 Information Systems

The information system as an amendment to driver assistance technology
contributes to possible accident reduction to a large extent. This means not
only visible roads but uniform, understandable, and clear traffic signs and the
integration of the driver and vehicle into an information and communication
system. Figure 6.33 shows the introduction of this type of support system
for the future [6- 131.
With navigation systems supported by traffic data, the driver can identify
traffic jams much earlier and thus avoid them. In the future, it also will
be possible to inform the driver about direct events, such as fog, icy roads,
accidents, and stopped traffic in a critical curve in the road. Other features
include "hands-free'' mobile phones, emergency calls, and services such as
e-mail or fax, naturally not while at the driver's seat or during driving. On a
national and international basis, much research is being done to improve these
systems and to define standards that allow the function, even if we change
the border from one country to another. These programs are performed on
an international basis.
In Europe, these programs include the following:
Information Society Technologies (Europe) (IST) (user friendly in the

information society)
The Intelligent Car Initiative "Raising Awareness of ICT for Smarter, Safer
and Cleaner Vehicles" [6-381
ISIS (Information Society Initiative in Standardization)
European Space Agency (ESA), new-generation spaceborne global posi-

tioning system (GPS), navigation, and mobile services
ERTICO, reliable travel and information; reliable and user-friendly public

transport; traffic management ensures efficient use of transport; faster
emergency services save lives; easy cash-less payment; high-quality and
cost-effective freight and fleet services; driver assistance systems for safer
vehicles and safer roads; and beyond-the automated highway
Programs in the United States and Japan include the following:
SAE International (active), accident prevention
U.S. Department of Transportation, safety
IVHS, electronic toll industry, commercial vehicle operations, traffic

management, border crossing, parking (access control), transit
The international programs are supported by national research [6-391. With

more advanced communication systems, better integration into a total traffic
management system is possible. This will allow the following:
Accident Avoidance
Planning of the driving route, also from home
Commercial fleet management, optimal use of roads and traffic means

(e.g., cars, rail, buses)
The responsible traffic planner also has an opportunity to shape the traffic
flow. If we look further into the future, other technical features will be pos-
sible, such as:
Pre-crash determination
Automatic emergency braking
Steer-by-wire
Collision avoidance
Electronic co-pilot
Autonomous driving
It is not efficient enough to optimize only the vehicle. We also must make
the road more intelligent (see Figure 6.34) [6-40,6-411. Some items for this
area are the following:
Electronic Tall Collection

License Plate Reader
Hazardous Road
Inlell. Traffic Lights
Figure 6.34 Mobility with vision. (Source: Refi 6-41.)
65
Increased visibility
Warning in advance of hazardous roads
Vehicle counting and classification
Identification of traffic congestion
Obstacle and accident warning
Intelligent traffic light and speed adaptation
Limited tunnel access
The date of the introduction of autonomous driving cannot be determined

today. It might be appropriate to start this technique not in normal public traffic
but in special applications. Nevertheless, the other systems described will be
introduced into the field via a step-by-step approach. The pre-conditions for
this are driver acceptance, optimization of the man-machine interface, and
achievement of an overall benefit.
6.5 References
6-1. Bundesministerium fur Verkehr, Bau und Wohnungswesen (Ed.),
Verkehr in Zahlen 2004/2005, Vol. 33, Deutscher Verkehrsverlag,
Hamburg, ISBN 3-87 154-314-4.
6-2. Busch, S., Schwarz, T., and Zobel, R. "Determination of Risk Factors
of Accident Causation," Institution of Mechanical Engineers Conference
on Vehicle Safety, May 28-29,2002, London, Volkswagen AG.
6-3. Behr GmbH & Co., Stuttgart. Klimatisierung [Air-conditioning], VDA-

Report, September 200 1, Frankfurt am Main, Germany.
6-4. Braess, H.-H. and Seiffert, U. Handbuch Kraftfahrzeugtechnik, Vieweg-

6-5. Richter, B. Schwerpunkte der Fahrdynarnik, Verlag TUV Rheinland

GmbH, Koln, 1990, ISBN 3-88585-772-3.
6-6. Becker, Axel, el al. "Maanahmen zur Verkurzung des Anhalteweges

in Notbremssituationen-das '30 m Auto,' " Proceedings, Verband der
Automobilindustrie (Association of the [German]Automobile Industry)
Accident Avoidance
Technical Congress, March 26-27, 2001, Bad Homburg v. d. Hohe,

Germany.
Peters, M. "Bedienkonzept im Fahrzeug-Das intuitive und adaptive

HMI," Forschungs- und Entwicklungszentrum-MMI-BMW AG,
Munchen, Germany.
Braess, H.-H. and Seiffert, U. Handbuch Krqftfahr~zez~gtechnik,

Vieweg-
National Highway Traffic Safety Administration (NHTSA) and Federal

Motor Vehicle Safety Standard (FMVSS), FMVSS 103, Windshield
Defrosting and Defogging Systems, Washington, DC, United States.

Wiping and Washing Systems, Washington, DC, United States.
Society of Automotive Engineers, SAE J902, "Passenger Car Wind-

shield Defrosting Systems," SAE International, Warrendale, PA, United
States.
Robert Bosch GmbH, Stuttgart. Product information, undated.
Specks, W. "Elektronikkonzepte fur zukunftige Fahrerassistenz-

systeme," Verband der Automobilindustrie (Association of the [German]
Automobile Industry) Technical Congress, March 26-27, 2001, Bad
Homburg v. d. Hohe, Germany.
Knoll, P. "Surround Sensing-Collision Warning Systems-Vehicle

Guidance," 01A2002, Bosch Group, Moglingen, Germany.
Hendrischk, W., et a[. "Beleuchtung," Handbuch Kraffahrzeug,

4 Auflage 2005, Vieweg-Verlag, Wiesbaden, Germany, ISBN 3-528-
331 14-3.
Eichhorn. U. "Assistenzsysteme fiir das Auto der Zukunft," Fortschritt-

Berichte VDI Series 12, No. 485, Diisseldorf, Germany, 2002.
6- 17. Robert Bosch GmbH (Ed.). Kraftfahrtechnisches Taschenbuch, 22nd

ed., VDI-Verlag, Diisseldorf, 1995, ISBN 3- 1 8-4 1 9 122-2.
6-18. Pape, E. "VW in 4 Motion," Proceedings of the Four-wheel Drive

Congress, February 14-15, 2002, Graz, Austria.
6-19. Zanker, W. "Neu in der S-Klasse: PRE-SAFE Bremse," June 2006,

[email protected].
6-20. Knoll, P. Surround Sensing and Sensor Data Fusion Technical Congress,
March 2002, Stuttgart, Germany.
6-2 1. Author's unpublished data. See also Zobel, R. "Analyse des realen
Unfallgeschehen Methoden und Prinzipien der VW-Unfallforschung,"
Proceedings of the conference "Ko11isionsschutz im Straflenverkehr,"
November 6-7, 1995, Haus der Technik, Essen, Germany.
6-22. Braess, H.-H. and Seiffert, U. Handbuch Kraftahrzeugtechnik, Vieweg-

6-23. Audi AG, Audi Q712006, Ingolstadt, Germany.
6-24. Robert Bosch GmbH (Ed.). Automotive Handbook, ISBN 0-89 283-
5 18-6.
6-25. Booz, Othmar, et a/. "Electro-Hydraulic Brake with the Focus on the
Electric Power Supply," Verband der Automobilindustrie (Association
of the [German]Automobile Industry) Technical Congress, March 2002,
Stuttgart, Germany.
6-26. Belschner, R., el a/. Brake by Wire Using a T T P K Communication

Network, VDI-Berichte 14 15, Electronic Systems for Vehicles, VDI-
Verlag, Diisseldorf, 1998, ISBN 0083-5560.
6-27. Gombert, B. "All 4 Engineers," July 2006, Vieweg-Verlag, Wiesbaden,

Germany.
Accident Avoidance
6-28. Schweigert, W., et al. "Das Fahnverk des neuen Passat," ATZ/MTZ
Special Edition, April 2005, Vieweg-Verlag, Wiesbaden, Germany.
6-29. Heifling,B. and Brandl, H.J. Subjektive Beurteilung des Fahrverhaltens,

Vogel-Verlag, Wurzburg, Germany, 2002, ISBN 3-8023-1906-6.
6-30. Analysis by the National Highway Traffic Safety Administration,

Insurance Institute for Highway Safety, and accident research teams
of vehicle manufacturers.
6-3 1. Berkner, S., et al. "Beeinflussung der Querdynamik vom Pkw durch
aktive Fahnverke, HDT." Conference on Driveability: Fahrkomfort,
FahrspaB, und Fahrsicherheit, Essen, June 2627,2001,
6-32. Huinink, H., et al. "Dynamische Interaktion Bremse-Reifen-StraBe,"

18 y Symposium, October 23-24, 1998, Bad Neuenahr, Germany; in:
Fortschritt-Berichte VDL, Series 12, No. 373, VDI-Verlag, Dusseldorf,
Germany, 1998.
6-33. "Besser im Griff," Automobil Industrie, May 2002, Germany.
6-34. Beru Aktuell, "The Beru RDKS," January 2002, Ludwigsburg,

Germany.
6-35. Binfet-Krull, M., et al. "Definition of SafetyIReliability Requirements

for Components of Electronic Vehicle Systems Like Steer by Wire," 4th
European Conference and Exhibition, Coventry, June 27-28, 2001.
6-36. Frank, D., et al. "Wo liegen die Grenzen der Fahrerassistenz?" Proceed-
ings, VDATechnical Congress, March 200 1, Bad Homburg, Germany,
Frankfurt 200 1.
6-37. Delphi Automotive Systems. Product information material during the

IAA 2001, Frankfurt, Germany.
6-38. Commission of the European Communities, February 2006, COM

(2006) 59 Final, On the Intelligent Car Initiative, "Raising Awareness
of ICT for Smarter, Safer and Cleaner Vehicles."
6-39. Kasties, G. "Verkehrstelematik: Staus miissen nicht sein," in Die

heimlichen Siegerbr-anchen,FAZ Buch Frankfurt, 2002, ISBN 3-89843-
008- 1.
6-40. Insurance Institute for Highway Safety annual status report, Vol. 37,
No. 5, May 4, 2002, Arlington, VA, United States.
6-41. Hoefflinger, B. Mobility with Vision, Institute for Microelectronics,

Stuttgart, Germany, 2001.
Biomechanics and
Occupant Simulation
7.1 Definition
For mechanical engineers, it might be surprising that the discipline of bio-
mechanics has become so important to develop vehicles by understanding
injury mechanisms. Biomechanics can be called the science that applies the
principles of mechanics to biological systems [7-11. It is not a new research
discipline. Galileo (1 564-1 642) and Harvey (1578-1658) worked in this field
during their time. Biomechanics includes in-depth studies on the behavior
of humans under internal and external forces, as well as applied engineering
work. Therefore, one could understand that many different disciplines are
included in biomechanics: engineering, epidemiology, traumatology, anatomy,
biology, and physiology.
In the last 50 years in the United States, Europe, and Japan, research activi-
ties in this area have increased. The necessary link between medical experts
and engineering was created because many organizations in various countries
openly exchange their research results at conferences such as the Experimental
Safety Vehicles (ESV), Stapp Car Crash, International Research Council on
the Biomechanics of Impact (IRCOBI), SAE International, and others, as
well as by personal contacts. Furthermore, large vehicle manufacturers and
insurance companies are supporting some of the work in this field.
An important contribution to knowledge about the resistance of humans against

a high-rate impact load was made by an American, Colonel John Stapp, who per-
sonally served as the first human to be decelerated from 632 mph (= 1000 k d h )
in 1.4 sec to zero. If we take a rectangular deceleration pulse, this corresponds
to approximately 20g. The annual Stapp Car Crash Conference therefore is
named after Colonel John Stapp. At the Eighth Stapp Conference in 1964
[7-21, the following appreciation to Colonel Stapp was made:
The Stapp Car Crash Conferences are named in honor ofcolonel

John Stapp, USAF (MC), who pioneered (and is stillpioneering;)
in establishing human zmpact tolerance levels. His historic rocket
sled rides at Holloman Air Force Base, New Mexico, in 1954, in
which he voluntarily subjected himself to up to 40g accelerations
while stoppingJFoma speed o f 632 miles per hour in 1.4 seconds,
still represent the best basis for quantzfiing hziman tolerance to
acceleration. In addition to his own dangerous volunteer work, he
has directed countless other safety research programs involving
hziman volunteers, animals, and cadavers. The equipment and
techniquesdeveloped under his guidance have become standard in
this research area and have contributed much to the advancement
of safety. The naming o f these confirences ajier Colonel Stapp
is aJitting tribute to (I man who has dedicated his life-even to
the point of risking it-to research aimed at increasing man b
chances of szirvival in adverse crash environments.
The conferences were initiated at the University of Minnesota

(Colonel Stapp b alma muter) zinder the able direction of Professor
James J. Ryan, another outstanding researcher in crash safety.
For,four years, the conferences were held at either the University
of Minnesota or an appropriate U S . Air Force base. Currently,
the conference rotates annually among four sponsors: The
Universityof Minnesota (196l), the UnitedstatesAir Force (1962),
the University of California at Los Angeles (IY63), and Wayne
State University (1964). The 1965 meeting will again be held
at the University of Minnesota on October 20, 21, and 22. The
proceedings of the conference are published in bound form and
will, it is hoped, become a valuable reference source. [7-21
Colonel Stapp is only one of the pioneers. Around the world, many others
are active in this type of research. In connection with vehicle safety of traffic
participants, the biomechanical results also are an instrument for determining
the biomechanic limits of humans. The results of biomechanic research led
to the definition of load limitations. From that, protection criteria are taken,
Biomechanics and Occupant Simulation
which should serve as limits that should not be exceeded. For this reason,
we first need to know what is happening in real-world accidents and which
injury mechanisms are important.
7.2 Injury Tolerance Limits
Injury tolerance limits describe items such as fractures, injuries of organs, and
other injuries. A classification is done via the Abbreviated Injury Scale (AIS)
and Overall Abbreviated Injury Scale (OAIS). With the AIS or OAIS, single
or total injuries are described. The data span a range from 0 to 6. Table 7.1.
shows the severity of injury rating versus the AIS. To more clearly illustrate
the relationship between injury description and the AIS, Table 7.2 provides
an interesting description.
TABLE 7.1
THE ABBREVIATED INJURY SCORE (AIS) [7-31
AIS Severity Code
No injury
Minor
Moderate
Serious
Severe
Critical
Maximum injury (virtually unsurvivable)
Unknown
The limits from the injury level depend on age, sex, anthropometrics, mass,
mass distribution, and specific conditions. This is why it is relatively com-
plicated to cover all traffic participants in accident simulation tests-from
vehicle occupants via cyclists to pedestrians-because we are unable to
carry out development tests with humans. Therefore, it can be observed that
over recent years, even more test devices or three-dimensional dummies or
computer models have been used by engineers.
TABLE 7.2
- AIS EXAMPLES BY BODY REGION [7-41
Abdomen and Extremities
Pelvic and
AIS Head Thorax Contents Spine Bony Pelvis
1 Headache or Single rib Abdominal wall; Acute strain Toe fracture
dizziness fracture superficial (no fracture or
laceration dislocation)
2 Unconscious 2-3 rib Spleen, kidney, Minor fracture Tibia, pelvis,
< I hr; linear fracture; or liver; without or patella;
fracture sternum laceration or any cord simple
fracture contusion involvement fracture
3 Unconscious 24 rib Spleen or Ruptured disc Knee
1-6 hr; fracture; kidney; major with nerve dislocation;
depressed 2-3 rib laceration root damage femur
fracture fracture with fracture
hemoth. or
pneumoth.
4 Unconscious 2 4 rib Liver; major Incomplete Amputation
6-24 hr; fracture with laceration cord or crush
open fracture hemoth. or syndrome above knee;
pneumoth.; pelvis crush
fail chest (closed)
5 Unconscious Aorta Kidney, liver, or Quadriplegia Pelvis crush
>24 hr; large laceration colon rupture (open)
hematoma (partial
(100 cc) transection)
7 . 3 External Injuries
A rating of cuts in faces was done by Professor L. Patrick of Wayne State
University. Because of the laminated windshield, seat belts, and airbags, this
type of injury is significantly reduced. The breaking of the skull due to impact
onto rigid parts was described in Ref. 7-5 as follows: The deceleration on the
head multiplied with the head mass results in a force that could break part of
the head. Figure 7.1 shows mechanical values for parts of the human body
and the injury level.
For example, the limits for the forehead are 80g, the nose 30g, and the chin
40g. As mentioned, this data could not be used directly if dummies were taken
for the analysis because dummies are not a precise duplication of humans.
Body Part Mechanical Variables Load Values

'otal Body ax max 40 ...809
2, 40 ...459, 160...220 ms
lrain ax a, ,ax loo...3009

WSU-curve wilh 60g. T>45 ms
1800...7500 radls2
;kull Fracture ax a, ,ax 80 ...3009 depending on the

size of the impact area
'orehead axmax 120...2009

F, 4000-6000 N
:ervical Spine ax mar l h r a x 30...409

ayma. tqorax 15...189
FK 1200...2600 N shear force
Omar rawam 80" ...100"
Omax r a m 80' ...90"
I I
'horax axmar 40 ...609, 1>3 ms
60g, 1<3 ms
'elvis-Femur FX 6400...12500 N force

application in the femur
A,,. 50..Jog (pelvic)
'ibia 2500...5000 N
150...210 Nm
120...170 Nm
Figure 7.1 Biornechanical limits on humans. (Source: Re$ 7-5.)

7 . 4 Internal Injuries
The injury mechanism for internal parts of the human body is much more
complicated to determine. The biggest problem is, without a doubt, the load
on the brain and the cervical vertebra. For the head, the tolerance level is the
limitation of the g-level in the anterior-posterior direction, with a value of
80g over a time period of more than 3 ms that should not be exceeded.
7.4.1 Concussion
A basic work on this subject was conducted by Wayne State University
under the supervision of Professor L. Patrick [7-61. In this experiment,
embalmed cadavers were used and impacted a rigid or padded flat plate.
Figure 7.2 shows the results of these tests.
It can be stated clearly that this is the basis for many rule-making decisions in
setting up the specified limits. If we analyze the curve shown in Figure 7.2,
it can be easily understood that there is a g-level time relationship. For this
reason, the severity index (SI) was developed, which was calculated by the
following formula:
Decelerallon
2400
I
m/s I
1600
1200
800 -- I
I
400
0
i
0 10 20 30
I
ms 50
Time
Figure 7.2 Patrick curve (scalefor the evaluation of loadings

on the human brain).
From a biomechanical viewpoint, a maximum level of 1000 should not be

exceeded for a unidirectional acceleration measurement. Later, new criteria
were established, which are known in the United States as head injury criteria
(HIC) and in Europe as head protection criteria (HPC):
where t is in seconds, a is the resultant acceleration measured in [g] in the

head, and t, and t2 are arbitrary time points. The HIC is calculated by an
iterative mathematical model in such a way that the measured acceleration
time function is achieving a maximum. To demonstrate the influence of the
shape of the acceleration time function, three different characteristic curves
are shown in Figure 7.3.
Although the maximum value of all three curves has the same level of IOOg,
the relevant HIC calculated has values of 246, 419, and 1000. The HICs
mentioned are always the maximum values for the specific deceleration time
function time t, and t2,as determined by an iterative calculation for all possible
t , and t2, insofar that a maximum level is achieved. In addition to the fact that
the HIC has only a limited information value (e.g., the HIC does not consider
the rotational influence and is not based on the resultant head acceleration),
we agree that this formula and the limit of 1000 [-I is used on a worldwide
basis to judge the performance of vehicles in accident simulation tests.
The rotational acceleration in the anterior-posterior direction was already

investigated by Fiala [7-71 in the late 1960s. His investigation showed that
with a brain mass of 1300 g, a value of 7500 rad/s2 should not be exceeded.
Resultant deceleration
HIC = 1000
=Oms
t2
= 10ms
a = 1oog
HIC = 246.4
t, = 4.3 ms
t2
= 10ms
a = 71.5g
0 5 ms 10
HIC = 41 8.661
= 1.6 ms
t2
= 8.3 ms
a = 82.8543g
Figure 7.3 Head injury criteria (HIC), as afi~nctiono f three dijferent

deceleration time curves.
7.4.2 Spinal Injuries

Another important field is the prevention of spinal injuries. Early in 1971,
Patrick and Mertz determined the torque moments for flexion and extension
via volunteers and human cadaver testing [7-81. Figure 7.4 shows some
head-neck responses. Table 7.3 shows some results about maximum torque
and forces related to the measurement with human volunteers [7-91.
2 0 10 20 30 40 50 60 70 80 S€
H e a d R o t a t i o n R e l a t i v e t o T o r s o (deg.)
0 20 40 60 80 100
Hcad R o t a t i o n R e l a t i v e t o T o r s o (deg.)
Figwe 7.4 Head-neck Ieesponse envelopes injexion and extension for the
loading phase. (Source: Re$ 7-8.)
TABLE 7.3
MAXIMUM STATIC FORCES AND BENDING TORQUES
DEVELOPED AT THE OCCIPITAL CONDYLES
BY HUMAN VOLUNTEERS [7-91
Bending Torque [Nm]
Forward flexion
Extension
Lateral flexion
Force [N]
Anterior-posterior (shear) 845
Posterior-anterior (shear) 845
Lateral shear 400
Axial tension 1134
Axial compression 1112
7.4.3 Chest Injuries

The chest is another area of critical injuries. 1t is evident that deformation
force, deformation speed, and the deformation itself have an influence on chest
injuries. Figure 7.5 shows force deflection measurements on unembalmed
cadavers [7- 1 01.
Where the force level has an average value of approximately 3 kN, deformation
is in the range of 60 to 100 mm (2.4 to 4 in.). In lateral impacts, we found
the force deflection curve as shown in Figure 7.6 [7-1 I].
It also was determined that deflection alone was not enough to determine
possible injuries. This was why Viano established the viscous criteria (VC)
[7-121, with the results as illustrated in Figure 7.7.
0.0254 0.0508 0.0762 0.1016

Total Deflection [m]
Figwe 7.5 Dynamic force dejlection in,frontal impacts.

(Source: Re$ 7-10.)
6.810
I
Rlbs fractured
...*-. No ribs fractured
-
'Denotes impad
Penetration [m]
Figure 7.6 Dynamic,force deflection in lateral impacts with padded

armi*est-simulatingimpactor: (Source: Refi 7-1I .)
Crushing Injury
-
1 2 5 10 20 50 100
Velocity of Deformation (mls)
Figzire 7.7 Ranges o f validity for the viscozrs criteria (VC).

(Sozirce: Rej.' 7-12.)
7.5 Criteria in the Rule-Making Process

The basic requirements are defined in FMVSS 208 [7-131, the EEC directive
for frontal impacts [7-141, and the EEC directive for lateral impacts [7-151.
7.5.1 Head Protection

Head protection criteria, HIC or HPC, should not exceed 1000 or 700 and
are calculated as follows:
a. 1. For any two points in time, t, and t2, during the event which are
separated by not more than a 36-ms time interval, and where t, is less
than t2, the head injury criterion (HIC36) shall be determined using
the resultant head acceleration at the center of gravity of the dummy
head, a , expressed as a multiple of g (the acceleration of gravity),
and shall be calculated using the expression
2. The maximum calculated HIC36 value should not exceed 1000.
b. 1. For any two points in time, t l and t2, during the event which are
separated by not more than a 36-ms time interval, and where tl is less
than t2, the head injury criterion (HICIS)shall be determined using
the resultant head acceleration at the center of gravity of the dummy
head, a,expressed as a multiple of g (the acceleration of gravity),
and shall be calculated using the expression
2. The maximum calculated HICISvalue should not exceed 700.
7.5.2 Chest Protection

Chest Deflection.
VC = viscous criteria <I mlsec
a. Compressive deflection of the sternum relative to the spine shall

not exceed 76 mm (3 in.).
Lateral rip deflection <42 mm (1.65 in.)
a. Compressive deflection of the sternum relative to the spine shall

not exceed 63 mm (2.5 in.).
b. Compressive deflection of sternum relative to the spine shall not

exceed 50 mm (2 in.) (European requirement).
7.5.3 Neck Injury

When measuring neck injury, each of the following criteria shall be met:
The shear force (F,), axial force (F,), and bending moment (My) shall
be measured by the dummy upper neck load cell for the duration of the
crash event as specified. Shear force, axial force, and bending moment
shall be filtered for Nij purposes at SAE 52 1111 rev. March 95 Channel
Frequency Class 600.
During the event, the axial force (F,) can be either in tension or compres-
sion, while the occipital condyle bending moment (Mocy)can be either
in flexion or extension. This results in four possible loading conditions
for Nij.: tension-extension (N,,), tension-flexion (Nt3, compression-
extension (N,,), or compression-flexion (Ncf).
When calculating Nij using the equation in (4) below, the critical values,
F,, and My,, are:
F,, = 6806 N (1530 Ibf) when F, is in tension
F,, = 6 160 N (1385 Ibf) when F, is in compression
Myc= 3 10 Nm (229 lbf-ft) when a flexion moment exists at the occipital

condyle
M = 135 Nm (100 lbf-ft) when an extension moment exists at the

YC.
occipital condyle
At each point in time, only one of the four loading conditions occurs.
The Nij value corresponding to that loading condition is computed, and
the three remaining loading modes shall be considered a value of zero.
The expression for calculating each Nij loading condition is given by
None of the four Nij values shall exceed 1.0 at any time during the event.
Peak tension force (F,), measured at the upper neck load cell, shall not
exceed 4170 N (937 lbf) at any time.
Peak compression force (F,), measured at the upper neck load cell, shall
not exceed 4000 N (899 lbf) at any time.
Unless otherwise indicated, instrumentation for data acquisition, data

channel frequency class, and moment calculations are the same as given
for 49 CFR Part 572, Subpart E Hybrid I11 test dummy.
Neck rearward torsion moment should not exceed 57 Nm (requirement

for Europe).
Force in longitudinal direction (requirement for Europe), F,, < (1.1-3.1

kN) as f(t) (see Figure 7.8).
10. Neck shear force (requirement for Europe) Fshear< (I .l-3.1) kN as f(t)
(see Figure 7.9).
Time [ms]
Figure 7.8 Neck tension load as a,finction o f time.

E
e,
4
2
0 3
LL
L
m
22
V)
Y
$ 1
Z
0
0 25 30 35 45 50 60
Time [ms]
Figure 7.9 Neck shear,force as aafi~nctionof time.
7.5.4 Pegonnance Criteria for the Rule-Making

Process
Of many possible values, the following limits were established in the rule-
making process.
7.5.4.1 Chest
Resultant chest acceleration <60g (>msec)
TTI = Thoracic Trauma Index 4 5 g (four-doors)

<90g (two-doors)
TTI = 0.5 x (RIBY + T12Y)
RIBY = Maximum absolute value of lateral acceleration in g's of the

fourth and eighth rib in the struck side
T12Y = Maximum absolute value of lateral acceleration in g's of the

twelfth thoracic vertebra after filtering of the acceleration signal
Force < 8000 kN

7.5.4.2 Pelvic
Resultant acceleration <130g
Force abdomen <2.5 kN
Force symphysis <10 kN
7.5.4.3 Leg and Knee

Upper Leg:
Force limit in frontal impacts <10,000 N (requirement FMVSS

208)
Shear load in the knee joint <5000 N
Knee dislocation <15 mm (requirement for Europe)
Force in the femur in the longitudinal direction (requirement for

Europe), Flong< (7.6-9 kN) as f(t) (see Figure 7.10)
Lower Leg:
Force in the tibia (compression force criteria) (requirement for Europe)

F~ong< 8 kN
Tibia index TI measured at the top and bottom of each tibia must not
exceed 1.3 at either location (requirement for Europe)
The tibia index is calculated on the basis of the bending moments

(M, and My) by the following expression:
where
M~ = Bending moment about the x-axis

0 10 20 30 40 50 60
Time [ms]
Figure 7.10 Upper leg force as a function of tiine.
= Bending moment about the y-axis

MY
(Mc)R = Critical bending moment and shall be taken to be 225 Nm
FZ = Compressive axial force in the z direction
(F) = Critical compressive force in the z direction and shall be

taken to be 35.9 kN
The tibia index is calculated for the top and bottom of each tibia; however,
F, may be measured at either location. The value obtained is used for the top
and bottom TI calculations. Moments M, and My are measured separately
at both locations.
Due to the requirements already defined and because of improved restraint

systems, especially the combination of seat belts and airbags, the degree of
protection of the human body has increased. This means that the knee and
the lower leg area have become a relatively high priority. For the feet of the
front occupant, it is important to prevent too high a bending-flexion around
the y-axis.
7.5.4.4 Pedestrian Protection

A special case are the limits proposed for the protection of pedestrians, based
on European legislation [7- 161. Figure 7.1 1 shows the different requirements
of phase I1 of the European legislation still under discussion.
a. Limitation of the headform acceleration for two different head forms,

child and adult simulation
b. A simulation of a leg impacting the bumper of the vehicle front, where

the leg angle should not exceed 15", the shear distance of the knee should
be below 6 mm (0.23 in.), and the acceleration of the tibia is less than
l5Og.
c. The simulation of the upper leg against the bonnet of the front hood with
the requirement that the shear force should be lower than 4 kN and bend-
ing moment of the impactor less than 200 Nm.
7 . 6 Test Devices
Simulation of parts of the human body and the complete human is necessary
to test vehicle components and the vehicle. Although professional investiga-
tions with human volunteers have been used, these can be performed only in
situations where no injuries occur. From the experience of one of the authors
at an age of 32 years, it can be stated that the level of the crash test should
not be high enough to cause injuries. In a lateral car-to-car crash with an
impact speed of 34 km/h (2 1 mph), this corresponds to a change in velocity of 17
kmlh ( I 0.5 mph), restrained by a standard three-point seat belt, the following
observations could be made. The pulse frequency jumped from 105 to 175
during the impact phase, a small concussion could be observed, and muscle
pain occurred over the whole body. The Av of this test was below the injury
limit, but a slightly higher change of velocity already might have created
minor injuries. For the development of vehicles and vehicle components,
Adult Headform In1pclctor
Child Headform Impactor S l m e d : 11.1 m/s (40 k~nlh)
Speed: 11.1 m/s (40 kmlh) Mass: 4.8 kg
Mass: 2.5 kg
A
Upper Legform lrnpactor

Speed up to 11.1 mls (40
Mass up to 17 k g
Angle up to 47"
LBRL > 500mm

Upper Legform Impactor
Speed: 11.1 mls (40 kmlh)
Mass: 9.5 kg
Angle 0'
LBRL < 500mm

Lower Logform Impactor
Speed:ll .Imls (40 kmlh)
Mass: 13.4 kg
Angle 0"
Figzrre 7.11 Test requirementsfor pedestrian protection.

(Sotrrce: Rec 7-17.)
engineers must have reliable and reproducible results. Likewise, the test
or calculations must be performed in areas where injuries can occur. This
automatically prohibits the use of human volunteers as test specimens. For
this reason, many test devices are available and are used for the development
and evaluation of production cars.
7.6.1 Body Part Test Devices

For vehicles, component test devices are used in the following areas:
The simplest equipment is a measurement device to control the radius of the

outer parts of the vehicle. It must be greater than 3.2 mm (0.125 in.).
Another measurement is done by a head form, which defines the impact

zone at the dashboard and the vehicle interior.
7.6.1.1 Head Impact

Figure 7.12 shows a head impact test device that is used to measure the g-level
during an impact of the reduced pendulum mass of 6.8 kg (15 lb). The head
impact form can be equipped with a three-dimensional accelerometer. It also
can be used by attaching it with glue to foil, which shows the surface pressure
if contact with a vehicle interior part occurs during the crash.
The newer requirements of FMVSS 201 request that not only the dashboard
but also the vehicle interior be tested for head impact performance. This head
fom-specified in FMVSS 20 1, Part 572-is used in a free flying mode. The
requirements are defined in FMVSS 20 1 [7-181 and SAE J 92 1 [7-191.
7.6.1.2 Torso Impact

To check the steering wheel and assembly unit, a body of 36 kg (79 lb) in
accordance with SAE 944 [7-201 was used. The impact speed is 24 kmlh
(1 5 mph), and the test device represents a torso of a 50% male, as shown in
Figure 7.13. This test is due to the installation of airbags, and three-point
belts are no longer a development tool.
Deceleration
Velocitv
Time
Figure 7.12 Head impact test device.
92
Velocity Deformation Force
Time
Figure 7.13 Steering cohmn test
93
7.6.1.3 Pedestrian Accident Simulation Tests

The safety of pedestrians is a key issue in several regions of the world, espe-
cially in Europe and Japan. The following test devices are used for research,
legal activites, and consumer information:
Child head form, 2.5 kg, 130-mm diameter (5.5 lb, 5.1-in. diameter)
Adult head form, 4.8 kg, 165-mm diameter ( 10.5 lb, 6.5-in. diameter)
Hip impactor
Leg impactor
As shown in Figure 7.11, these devices should be used to evaluate the per-
formance of cars having frontal impacts with pedestrians.
7.6.2 Three-Dimensional Dummies

The simulation of humans to test performance in accident simulation tests is a
relatively new discipline. Until the mid- 196Os, occupant simulation was done
by sewn leather bags filled with sand. Since then, after equipping vehicles
with sophisticated safety features, test dummies have been designed to more
closely simulate human data, such as age, sex, and size. The measurement
technique required today also demands a much more complicated design
of dummy, as shown in Figure 7.14. Figure 7.15 shows the interior of the
Hybrid 111 dummy. The 50th percentile male dummy is specified under
Title 49 CFR Part 572, Subpart E.
In general, we also must take into consideration for the simulation of humans
by dummies the following parameters:
Degree of biofidelity, sensitivity to injury parameter

Repeatability
Reproducibility
Durability
Repairability
Calibration technique
Total cost
We likewise must differentiate, depending on the kind of investigation.

HEAD PCCELESI4ETEBS
Ax.
MKL L7;G. G-:HZNNEL

UPPEK HCLK L04D C C L L 1
CSLlB4AlIOI FIX7LRC YOOEL 1860
&= ap
Figure 7.14 Hybrid I// dummy, ~neasur*eernen~

possibilities [7-2.21.
Figure 7.15 Design of a Hybrid 111 dummy.

7.6.2.1 FrontaVRear Collision and Rollover

In accident simulation tests (e.g., in frontal and rear collisions and in rollover),
the 5% female, 50% male, and 95% male from the Hybrid series and child
dummies are used. Figure 7.20 shows the dimensions of several dummies.
The dummies must undergo a relatively complicated calibration test series
before they are used in compliance tests. These dummies also are very
expensive. In 2006, the cost of a 50% male dummy without instrumentation
was approximately $30,000.
7.6.2.2 Lateral Impact

For lateral impact, two different 50% male dummies are in use-one for the
United States, and one for European regulations. These are the U.S. side
impact dummy (US-SID) and the European side impact dummy (Euro-SID).
Figure 7.16 shows the differences between the two dummies [7-211.
United States Side Impact Dummy. European Side Impact Dummy,

US-SID Euro-SID-I
Figzwe 7.16 Comparison o f US-SID and EZIYO-SID

side impact dzrmmies.
97
The differences generally would not to be too big of a problem. However,

because an identical car gives significant differences in the same accident,
simulation tests make product development complicated if a car manufacturer
is delivering its car to Europe and the United States. Design features that are
positive for one side impact dummy appear to be negative for the other. There
are two developments related to the Euro-SID. One is the Euro-SID-2 (ES-2),
which is used in the EURO NCAP. The other is the Euro-SID-2re with
rib extension [7-221. Figure 7.17 shows the design change. This dummy
is proposed to replace the SID-H3 in the FMVSS 201 and the US-SID in
the FMVSS 214 testing. For the FMVS 201 in a pole test, the SID-H3
dummy-a combination of the US-SID and the Hybrid 111 50th percentile
Figure 7.1 7 Ettro-SID-2re [7-221.

dummy-is used [7-221 (Figure 7.18). In the future, as mentioned, there

is some hope that we can use only one side impact dummy worldwide.
Figure 7.19 shows the status of the year 2006 for the World-SID design
[7-221.
78051-61XS
96-SIDH000i
Couw.siou Parts ~~-SIDHZOOS
78051.243
7.6.2.3 Child Dummies

Child dummies are a special case that encompasses the simulation of babies
up to 10-year-old children. Figure 7.20 shows some examples, in addition to
the total range of child and adult dummies [7-221. Child dummies are becoming
Figure 7.I 9 Prototype of the World-SID, a worldwide

harmonized side impact dumm,v.
increasingly important, not only for the evaluation of child restraints but in
connection with the performance of front airbags.
7.6.2.4 Rear Impact Dummy (Bio-RID)

Because of the large number of rear-end collisions in Sweden, a new anthropo-
morphic test device with special attention to the evaluation of possible neck
injuries in rear-end collisions was developed by the Denton company with
Chalmers University in Goteborg, Sweden. The Bio-RID dummy represents
a 50% male with an articulated thoraco-lumbar spine and neck made from
a composite material. The motion of the cervical vertebra is controlled by
cables that are attached to neck-muscle substitutes and dampers. Details are
shown in Figure 7.21 [7-221.
P
,- r4
3
ill
- .
+.
1'.
i r,
Figure 7.20 Duinnzy family. (Source Re$ 7-22)

Figure 7.21 Bio-RID dz~mmy,forrear impact applications.
7.6.2.5 Biofidelity Dummy

It also is often discussed whether the automobile industry needs a dummy
that more closely simulates human behavior. The answer to this question
depends on the purpose of the research analysis. For single research investi-
gations, this might be acceptable. However, for general use in development
and certification work, it certainly is not.
7.6.3 Human and Dummy Modeling

For several applications, human models and parts of the human body are
available. The models are used for accident simulation to reconstruct real-
world accidents, mathematical optimization for the design of the vehicle,
and biomechanical studies. Until now for certifications tests, the mathematical
simulations have not been used, although the results meanwhile are as good
as real-world accident simulation tests.
The simulation tools range from very simple models (one to two masses)
up to very sophisticated tools. The most frequently used multi-body crash
dummies are based on the MADYMO program. Figure 7.22 shows the wide
application field [7-1, 7-23].
MADYMO products range from a child dummy up to a 95% male adult

dummy. The multi-body dummy gives very good results, especially for
frontal impacts and using the restraint systems developed for this type of
accident. Because of the detailed analysis that is available and the fact that
the design engineer does not like to change computer software, finite element
method (FEM) simulation tools for dummies also are used [7-241, as shown
in Figure 7.23.
The FEM computer model simulations compare well with the real dummy
testing, especially in side impacts or in other simulation situations with a local
dummy loading. Both types of simulation tools are used. The MADYMO
system also gives good conformity to kinematics and to the influence of
the restraint system, and it requires less computer time compared to the FEM
system.
For more sophisticated investigations, a human body model is very expensive.

Application fields are new vehicle systems that are not in production; they can
simulate human parts such as the chest, neck, head, and feet. The simulation
model therefore is complicated because of the behavior of humans. Geometric
design data from RAMSIS were used to design a series of human models of
different sizes. Figure 7.24 shows multi-body human models representing a
midsize male (left) and small female (right).
Vehicle manufacturers and large supplier companies use these human models
for many of their analyses. Typical questions arise in areas where legislation
is not available, but where the development engineer needs some support in
the safety design.
Figure 7.22 Multi-body dummies for- simulation tests [7-231.

Figwe 7.23 Finite element dz~mmy.
Figure 7.24 Simulation tools genemted by the RAMSIS model.

Left: midsize male. Right: small female.
References
Wisman, J.S.H.M., eta/. "Injury Biomechanics" (4 J 610), Proceedings
of the Conference on Biomechanics, 3rd ed., Technical University of
Eindhoven, The Netherlands, 2000.
Proceedings of the 8th Stapp Car Crash Conference, Detroit, 1964.
"The Abbreviated Injury Scale-1990 revision," Association for

the Advancement of Automotive Medicine, Des Plaines, IL, United
States.
Pike, J. Az~tomotiveSafety, Society ofAutomotive Engineers, Warren-

dale, PA, United States, 1990.
Swaeringen, J.J. "Tolerance of the Human Face to Crash Impact," Die

Widerstandsfahigkeit des menschlichen Gesichtes gegen StoDe bei
Unfallen, Federal Aviation Agency, July 1965, Report No. AM65-20,
Oklahoma City, OK, United States.
Patrick, L. "Human Tolerance to Impact-Basis for Safety Design,"

SAE Paper No. 650 171, Society ofAutomotive Engineers, Warrendale,
PA.
Fiala, E., et al. "Verletzungsmechanik der Halswirbelsaule, Forschungs-

bericht der Technischen Universitat, Berlin 1970.
Mertz, J.M., et al. "Strength and Response of the Human Neck,"

Proceedings of the 15th Stapp Car Crash Conference, Coronado, CA,
1971.
SAE Information Report, "Human Tolerance to Impact Conditions as

Related to Motor Vehicle Design," Society of Automotive Engineers,
Warrendale, PA, United States, 1986.
7-10. Kroell, C.K. "Thoracic Response to Blunt Fronted Loading," in The

Human Thorax, Anatomy, Injury, and Biomechanics, P-67, Society of
Automotive Engineers, Warrendale, PA, United States, 1976.
7- 11. Stalnaker, R.L., Roberts, V.L., and Mc Elhaney, J.H. "Side Impact
Tolerances to Blunt Trauma," 17th Stapp Car Crash Conference, pp.
377408, Society ofAutomotive Engineers, New York, United States,
1973.
7- 12. Viano, D.C. "A Viscous Tolerance Criteria for Soft Tissue Injury Assess-
ment," J, Biomechanics, 2 1 15, 1988, pp. 3877399.
7-13. U.S. Department of Transportation, National Highway Traffic Safety

Administration, (NHTSA), and Federal Motor Vehicle Safety Standard
(FMVSS), FMVSS 208, 49 CFR Parts 552,571,585, and 595 Occupant
Crash Protection, Washington, DC, United States.
7-14. European Parliament and Council on the protection of occupants of

motor vehicles in the event of a frontal impact and amending Directive
701156/EEC, Brussels.
7-15. European Parliament and Council on the protection of occupants of

motor vehicles in the event of lateral impacts and amending Directive
701156/EEC, Brussels.
7-16. European Parliament and Council draft relating to the protection of

pedestrians and other road users in the event of a collision with a motor
vehicle, Brussels.
7-17. Friesen, F., et al. "Optimierung von Fahrzeugen hinsichtlich des

Beinaufpralltests," ATZ, May 2002, Vieweg-Verlag, Wiesbaden.
7-18. National Highway Traffic Safety Administration (NHTSA) and Fed-

eral Motor Vehicle Safety Standard (FMVSS), FMVSS 201, Occupant
Protection in Interior Impact, National Highway Traffic Safety Admin-
istration, Washington, DC, United States.
7- 19. U.S. Department of Transportation, National Highway Traffic Safety

Administration. Laboratory Test Procedure for FMVSS 20 1 "Occupant
Protection in Interior Impact Upper Interior Head lmpact Protection,"
Washington, DC, United States. Also Test Procedure in Accordance
with SAE 592 1 "Motor Vehicle Instrument Panel Laboratory Impact Test
Procedure," Society ofAutomotive Engineers, Warrendale, PA, United

States. Also SAE J 977, Instrumentation for Laboratory Impact Tests,
Society of Automotive Engineers, Warrendale, PA, United States.
7-20. Test Procedure in Accordance with SAE 944a, Society of Automotive

Engineers, Warrendale, PA, United States.
7-21. First Technology Safety Systems, Inc. Product description, Plymouth,

United States, www.ftss.com.
7-22. Denton ATD, Inc., 2006, product description, United States, www.
radenton.com.
7-23. TNO Automotive, www.tass-safe.com, MADYMO Model Manual,

Version 6.3, December, 2005.
7-24. Meywerk, M., et al. "Aspects of Optimization and Parameter Studies:

Case Studies from Automotive Industry," LMS-Conference for Physical
and Virtual Prototyping, Paris, France, 200 1.
Vehicle Body
8.1 General
The body-in-white and vehicle interior features contribute the most to vehicle
safety, especially in the areas of reduction of low-speed damage and to the
occupants of the vehicle in an accident. The body-in-white of a vehicle is not
primarily designed to mitigate injuries but to carry the vehicle components
(e.g., the powertrain and chassis) and the occupants of the vehicle. The first
patent in the field of vehicle body design related to safety was published in
October 1952 [8-11 by Bela BarCny. In his patent, he described how the struc-
tural strength should be greatest in the vehicle compartment and that the front
and rear of the vehicle should be less resistant to crushing and be capable of
absorbing energy during a crash. Figure 8.1 shows the differences. The left
side of the figure is a conventional design of that time; on the right side, the
different stiffnesses in the front and rear end of the vehicle are compared to
that of the passenger compartment.
Figure 8.1 Patent drawing by Bela Bare'nyi.

(Source: Re$ 8-1.)
With a weight of approximately 250 kg (551 lb) in an average subcompact

car design, the body-in-white represents slightly less than one-quarter of the
total vehicle weight. Realizing the increased vehicle requirements that have
occurred over the years, this low weight is a big improvement compared to
the past. The car occupants gained more interior length for room and other
advantages such as lower vibration and higher body stiffness of the vehicle.
Figure 8.2 shows the increase in torsional stiffness from one model generation
to the other for a compact car. Compact cars also have achieved a longer life
durability and low corrosion over a lifetime of more than 10 years.
Natural freauencv bodv-inwhite

Natural Cequency himmed body
/
Bending
- 5Hz
Predecessor A4 ncw
static torsional stlifness
Predecessor A4 new
Predecessor A4 ncw
Figure 8.2 comparison of the stiffness of a compact car for two

generations o f cars. (Souire: Audi AG.)
With respect to accident distribution, Figure 8.3 lists information for European
vehicles in which vehicle exterior body parts are involved in crashes [8-21.
Independent of the fact that in most accidents, the front of the vehicle is
involved, all areas of the vehicle body could be involved in accidents.
The design of the body-in-white has changed during the last few years.
Because of the necessity to minimize an increase in weight, regardless of the
new requirements, each single part of the body-in-white was analyzed for
Vehicle Body
angular
7.2% 17.2% 19.0% front
back
side angular front
sidc sidc
Figure 8.3 Statistical analysis of distribution of car accidents in Europe.
weight reduction and increases in performance. The following examples show

different and successful designs. The Volkswagen Beetle as built until 1974
had a rigid underbody with a stiff front end, which also carried the front axle.
Especially in car-to-car crashes, the design provided a good survival space,
despite the low weight of the Beetle. Figure 8.4 illustrates the design.
Most new cars use the self-carrying body-in-white design. As an example of

vehicles with a high production volume, the basic layout of the 2005 Volks-
wagen Passat [8-31 is shown in Figure 8.5.
Via a stable front cross-bar behind the bumper, the two longitudinal beams
and the upper fender transmit the forces in a crash to the middle tunnel and
the A-pillar at the height of the floor panel. The doors, including the cross-
bar for side impact protection, transmit forces to the rear of the vehicle. For
the optimization of the longitudinal beams, many design parameters can be
used. In addition to the optimum geometric configuration of the previously
mentioned reinforcement at the place of the highest bending moment, some
cars use longitudinal beams that are tailored metal blanks of different thick-
nesses to ensure that the beams do not bend but fold. In the frontal area,
these beams have a thickness of 1.5 mm (0.06 in.); the thickness in the rear is
2.5 mm (0.1 in.). To achieve a high energy absorption deformation, elements
Figure 8.4 Front end of the Volkswagen Beetle as of 1973.

(Sozwce: Volkswagen Safety Report, 1974.)
with a high-volume specific energy-absorption capability are required. The

deformation force time function should be kept as constant as possible.
Researchers have performed many studies in the field of collapse bulging.
Recent investigations found that an "inclined deformation" is even better.
The fuel tank must be placed in a well-protected area. Side and rear-end
impacts should not disturb the fuel tank system nor the interior integrity of
the passenger compartment. For rollover protection, the A- and B-pillars are
reinforced. If the occupants of the vehicle are wearing seat belts, these pillars
provide a good chance of survival in most rollover accidents.
Another interesting design is the DaimlerChrysler A-Class car [8-41. Figure 8.6
shows the basic layout of this vehicle.
Vehicle Body
Figure 8.5 Body strzrcture of a compact car (Sozwce: Re$ 8-3.)
Figure 8.6 The structz~r~al

design of the DaimlerChrysIer A-Class cau:
Although the A-Class design fulfills the basic requirements in its class, the
car is different because the front longitudinal beams are horizontal and are
designed without a geometric "S" shape. Because bending beams are not
absorbing enough energy, this alternative is one positive design feature. The
other feature is the capability that in a frontal crash, the engine and transmis-
sion should glide under the underbody, thus allowing a higher crash length with
respect to free deformation during frontal collisions. It also is one feature for
a positive design with respect to compatibility in car-to-car crashes because
the engine and transmission mass may not be harmful to the other colliding
vehicle in some specific types of accidents.
Another interesting design is the aluminum space frame technology of the

Audi A8 and A2. The design and production technique for the A8 as of today
is more advanced compared to the A8 of the first generation [8-51. More
welded aluminum structures are used instead of cast knots. Figure 8.7 shows
the specific design. Both vehicles could optimize the crash capability due to
the space frame technology. Other techniques are used, for example, for the
Aston Martin Vanquish, where many body parts are glued together.
Figure 8.7 The alz~minumspaceframe of the Audi A8.
114
Vehicle Body
For convertible and roadster vehicles, additional features are necessary. One
example used for the Mercedes SL is described as follows [8-71:
High occupant cell with multi-layer beams
Composite body floor with optimized material thickness including diagonal,

longitudinal, and transverse beams
Increase in the use of high-strength steel from 19% (predecessor) to

33%
Optimized wheel
Installation of a multi-piece elliptical front wall with a cross member
Additional beams in the upper front structure
Bending-resistant B-pillar with lateral longitudinal beams
Doors with reinforcement
Integral seat with stiff cushion seat-back frame
Reinforced A-pillar
Automatic rollover bar
Many other available designs are on the road, such as sport utility vehicles
(SUVs), multi-material concepts used in some sports cars, and one-box
designs for multi-purpose vehicles. However, these fit more or less into the
previously demonstrated concepts. As Figure 8.8 demonstrates, one will find
a high number of legal requirements for the vehicle body.
As mentioned, the body-in-white has many more duties. Figure 8.9 demon-
strates the numerous different forces that must be taken into consideration
by a good design over several 100,000-km (62,305-mi) driving distances and
more than 10 years of lifetime.
FMYGS216
Interior ,
FIIlVSS201,ZQZ 2W.
m4.205207.213.225
I {Headrests
Instrument panel , E C F - R ~ Z .I B . ~ ~ . Z I . A A I
I
FMYSSZM ECE-R 17.25 EG 7 m z
I MYSS 201 ECE-R ?I. 32.33 EG 7 1 1 0 \ ', I 1 Rollover
WV6S203.2W. 205,208.2OQ. 710.112

Bumper Side impact Seats
:E-R 12, 14, 16.3. 9r
,,I FUVSS
FMVSS21S bCL-R42 20' 2% 206.714
Steering ECE-Rl'. %
FMVSS 201. W 7
ECE-R 16.17.21.44
I G 0627 €67 6 1 0 . /4/408
wheel
rMVSS 203.2M
ECt-R 12
EG nnw
Figure 8.8 Legal requil-ements around the vehicle body

for mitigation of'injuries.
ear
In SY~COIIIP~CI
6.200 Nm
Figure 8.9 Externa1,forces to the body-in-white. Note: Nzlmer*ical values

in German spelling (7.000 N = 7,000 N in American spelling).
Vehicle Body
8.2 Low-Speed Impact

Several requirements are defined for the testing of vehicle integrity at low-
speed impacts and the performance of the vehicle in relation to repair costs at
high-impact speeds. Figure 8.10 provides an overview ofthe requirements in
the front of the vehicle, and Figure 8.1 1 provides an overview of the require-
ments for the rear of the vehicle.
The results are published in the consumer information systems and by insur-
ance companies. These results are used as the basis for vehicle damage
assessment of the relevant vehicle, as well as for the basis for insurance costs
Barrier I
Offset crash 40% overlap

lJ[overlap] = 0.4 . B
B is the width of the vehicle without exterior mirrors
Impact steering wheel side
v F = 1 5 k p h + l kph
Barrier clearly higher than the front of the vehicle
R = l5Ornm
No contact between the front of the vehicle and the wall next to the barrier
Vehicle occupied by one dummy, 50th percentile male, driver's position,
restrained
Fuel tank full of gasoline or diesel, water permissible
Suspension alignment check before and after crash test
Vehicle driveable and coasting before impact
Figure 8.10 Repair cost test front o f the vehicle).

Offset crash 40% overlap

11 overlap = 0.4 . B
B is the width of the vehicle without exterior mirrors
Impact on steering wheel side
StW is the moving barrier, coasting before impact
mStW = 1000 kg
Wheelbase dr 1 1.5 m
A is the width of the moving barrier 2 1.2 m
S is the center of gravity of the barrier in the center plant
H is the barrier height 700 mm
H is the lower edge barrier 20 mm
R = l5Omm
r=50mm
vStW = 15 kph + 1 kph
vF = 0 kph, brakes applied
No contact between the front of the vehicle and the wall next to the barrier
Vehicle occupied by one dummy, 50th percentile male, driver's position, belted
Fuel tank filled with gasoline or diesel, water permissible
Suspension alignment check before and after crash test
Vehicle ready to drive
Figure 8.11 Repair cost test (rear o f the vehicle).
Figure 8.12 shows a good example of a positive design in low-speed

impacts.
Related to repair costs at high speeds are not only the repair methods and
associated costs but the costs of the spare parts that also influence the rating.
In addition to the design, the vehicle manufacturer has other possibilities (e.g.,
low prices on spare parts) to reduce insurance premiums.
Vehicle Body
Displacement [mm]
Figure 8.12 Design feature,for good performance in low-speed impacts
(repair cost). (Source: ReJ: 8-9.)
8.3 Vehicle Body Test Without the Simulation

of Car Occupants
In the real-world crash environment, the vehicle occupant plays the major
role in combination with the vehicle related to occupant protection. However,
many tests involve the vehicle body without the use of three-dimensional
dummies.
8.3.1 Quasi-Static Test Requirement

8.3.1.1 Seat and Seat-Belt Anchorage Point Tests
If the lower inner anchorage point is mounted at the seat, which usually is the
case for the seat-belt latch, the seat and the seat-belt anchorage points are
tested simultaneously. Figure 8.13 shows a typical test configuration.
Figure 8.13 Seat-belt anchorage test.
By using rigid body blocks, the pulling force is applied equally to each anchor-
age point. (The test is done in accordance with FMVSS 210 [8-101.) For
each occupant seat, the resisting force should be greater than 14,000 N. The
upper anchorage point with the variable height adjustment and the latch at
the seat require special attention. For example, the reinforcement plate at the
B-pillar should not be stiff to prevent the outer B-pillar metal from being cut.
The belt latch at the seat also is important because the forces are transmitted
via movable metal parts to the stiffer portion of the underbody or the vehicle
middle tunnel. The reason for this is that in a standard seat design, the seat
cannot absorb the high forces from the seat-belt anchorage points. The seat
itself must resist more than 20g over a period of more than 30 ms. To transmit
the force to the middle tunnel, a serrated seat rail is used. Because of the load-
ing during an accident, the vertical component of the seat-belt pulling force
snapped the seat-belt anchorage point firmly into this special design element.
Figure 8.14 shows the design of such a solution (serrated strip).
Vehicle Body
Figu1.e 8.14 Design of the seat track mount and the seat track.
Seats in which the upper anchorage point also is mounted at the seat back
must have reinforcements installed in the seat back to absorb the forces and
moments. This often requires more weight and costs, and thus such a solution
is installed only in special models such as convertibles.
8.3.1.2 Roof Strength

For the evaluation of roof strength in accordance with FMVSS 2 16 [8-1 I], a
steel plate is used. This plate is inclined by 25" in relation to the horizontal
vehicle longitudinal plane and 5" to the front. Figure 8.15 shows the test
configuration.
The legal requirement requests that with a force that is 1.5 times the vehicle
curb weight (but not greater than 2267 kg [4998 lb]), the deformation mea-
sured perpendicular to the test plate should not exceed 12.7 cm (5 in.). This
requirement also must be met in convertibles. Because of the test configura-
tion, the greatest resistance is created by reinforced A-pillars, which together
with the glued-in windshield and the roof can provide sufficient survival
space in case of a rollover accident. In the future, a higher deformation
force will be requested: 2.5 times the vehicle curb weight. Also, instead of
a maximum deformation and force, a survival space for the occupant might
be the criterion.
0 50 100 mm 150
Figure 8.15 Roof test.
8.3.1.3 Vehicle Side Structure

In addition to the dynamic tests, the side structure of the vehicle is tested in a
quasi-static test, in accordance with FMVSS 214 [8-121. This test uses a half
cylinder that is pushed perpendicular to the longitudinal axis of the vehicle
into the door of the tested vehicle. The lower part of the cylinder has a height
Vehicle Body
of 12.7 cm (5 in.) above the lower part of the door sill. The cylinder has a
diameter of 30.5 cm (12 in.). The height is set so that the upper part of the
cylinder is at least 12.7 cm (5 in.) higher than the lowest part of the bottom
edge of the side windows. Figure 8.16 shows the test arrangement and the
Figure 8.16 Quasi-static lateral test.

intrusion force as a function of deformation length of the door. Crush resis-

tance is defined for three zones:
1. Initial: Not less than 2250 lb = 1010 N over a distance of 6 in.

2. Intermediate: Not less than 3500 lb = 15,570 N over a distance of
12 in.
3. Peak crush resistance: Not less than twice the curb weight or 7000 lb =
31,140 N
The lower curve describes the result for a door without a side beam, whereas
the upper curve describes the result for a door with a side beam. During such
tests, the reachable maximum force is limited by the force transmission of
the door, including the side beam, to the hinges and the door latch. In some
new designs, the lower part of the door also is anchored with the door sill.
Because the maximum force is limited by the resistance of the A- and B-pillars
and the lower anchorage mechanism, the maximum force is approximately
the same for the door without the side beam. The real positive effect occurs
during the first 240 mm (9.45 in.) of deformation. This means that especially
at the beginning of an impact with another car, the resistance on the side of
the impacted vehicle should be as high as possible.
Some research engineers believe that the side of the vehicle should be built
similar to a board frame, as shown in Figure 8.17 [8-131. If the side of the
struck vehicle and the front of the impacting vehicle are designed to be
w.4'
Figure 8.I 7 Board,frame layozlt.
Vehicle Body
compatible, a slide-away effect could reduce the potential of intrusion and

the risk of injuries.
Because of new dynamic tests with modern dummies, the question gener-
ally arises from time to time whether these quasi-static tests are of value in
improving safety. Although the tests are far from real-world accident per-
formance, we believe that these minimum requirements are a good starting
point for the designer to lay out the vehicle structure.
References
Barinyi, Bela. Kraftfahrzeug, insbesondere zur Beforderung von Per-
sonen, Deutsches Patentamt Nr. 854 157 30, October 1952.
Furstenberg, K.Ch., et al. "Development of a Pre-Crash Sensorial

System-The CHAMELEON Project," in VDI-Berichte 1653, Vehicle
Concepts for the 2nd Century of Automotive Technology, ed. by
Verein Deutscher Ingenieure, VDI-Verlag, Dusseldorf, Germany,
2001, pp. 289-3 10.
Hillmann, J., et al. "Der neue Passat," Vieweg-Verlag, April 2005.
Schoneburg, R., et al. "Weiterentwicklung eines bewahrten Sicherheits-

konzepts-Die neue Mercedes-Benz-A-Klasse," ATZ Special Edition,
October 2004, Vieweg-Verlag, Wiesbaden, Germany.
Scheurich, H., et al. "Der Audi A8-die neue Sportlichkeit im

Luxussegment," ATZ Special Edition, August 2002, Vieweg-Verlag,
Wiesbaden, Germany.
"Mercedes SL 2001,"ATZ/MTZ Extra, October 2001, Vieweg-Verlag.
Anselm, D. The Passenger Car Body, Society ofAutomotive Engineers,

Warrendale, PA, 2000.
8-9. Haberer, K.H. "Das Stofifangersystem des AUDI A4, Der Neue AUDI
A4," ATZ Special Edition, Vieweg-Verlag, Wiesbaden, November
2000.
8- 10. National Highway Traffic Safety Administration (NHTSA) and Fed-

eral Motor Vehicle Safety Standard (FMVSS), FMVSS 210, Seat Belt
Assembly Anchorages, National Highway Traffic Safety Administra-
tion, Washington, DC, United States.
8- 11. National Highway Traffic Safety Administration (NHTSA) and Federal

Motor Vehicle Safety Standard (FMVSS), FMVSS 216, Roof Crush
Resistance, National Highway Traffic Safety Administration, Washing-
ton, DC, United States.
8- 12. National Highway Traffic Safety Administration (NHTSA) and Federal

Motor Vehicle Safety Standard (FMVSS), FMVSS 214, Side lmpact
Protection, National Highway Traffic Safety Administration, Washing-
8- 13. Schimmelpfennig, K.-H. Unpublished data, Miinster, Germany.

Dvnamic Vehicle
Simulation Tests
9.1 Frontal Collisions

The frontal collision was the first type of accident that was analyzed in detail
for occupant protection. In addition to the previously mentioned low-speed
tests, impact speeds of up to 64 kmlh (40 mph) are being used today. The
obstacles vary from a rigid fixed wall, other vehicles, a pole, and a tree simula-
tion to various deformable barriers. The basic physical elements of a frontal
collision are explained by an analysis of a frontal collision against a fixed
barrier. The kinetic energy must be absorbed by the vehicle and by some
elastic rebound. Figure 9.1 shows the deceleration s, the velocity change S,
and the deformation length s as a function of time.
The rebound, visible by the negative velocity, shows that for this type of
accident simulation, the elastic rate is approximately 10%. This means
that the change in velocity for an impact against this fixed barrier at 50 km/h
(31 mph) is approximately 55 kmlh (34 mph). For a quick analysis, we
can assume a non-elastic impact performance. The formulas describe
the following:
SV
= Deceleration of the vehicle as function of time f(t)
Sv
= Velocity of the vehicle during the impact as f(t)
Sv = Deformation of the vehicle during the impact as f(t)

F, F = Deformation force, average deformation force
vi = Impact velocity
Av = Change in velocity
Impact Velocity Vi
A
V i g Impact Velocity
Change in Velocity
Deformation
Velocity
S= f(f) Deformation
Time
Deceleration
Figure 9.1 Deceleration, velocity, and deformation as ajunction o f time.
If we further stipulate that the average deformation force remains constant

over time, then the following correlation is valid:
Dynamic Vehicle Simulation Tests
If we also replace the average deformation force through the expression

F = m, . a , then we find another interesting relationship:
From that formula, we can conclude that the deformation force could be
different from the car-to-car design. As vehicle mass increases, so does the
force, F = m . a . If the average deceleration in frontal barrier impacts is similar
among the cars on the road, then the deformation forces are mass dependent.
In reality, smaller cars have a shorter deformation length and thus have a
higher g-level during frontal impacts. The deformation length for cars at
an impact against a fixed barrier at 50 kmlh (3 1 mph) varies between 450 to
750 mm (18 to 29.5 in.). At 56 kmlh (35 mph), it reaches values of 550 to
850 mm (22 to 33.5 in.). Special cars such as micro-compact vehicles (i.e.,
smart electric vehicles) have smaller deformations, in the range of 300 of
350 mm (12 to 14 in.). Although no measurable lateral force component
occurs during a straight, frontal collision, we find many transverse forces in
other crash modes.
In addition to the acceleration measurement on the tested vehicle, some test

facilities measure the reaction force at the barrier. Figure 9.2 shows the
resultant force, measured as single forces with 72 segments in the front of the
barrier and summed to the total force, versus time and deformation [9- 11.
Tlme [ms]
Dsformation [mml
Figure 9.2 Measzwed,forces at a rigid barriel: (Source: Re$ 9-1.)
Figure 9.3 shows an instrumented deformable crash barrier, which uses 64

independent triaxial load sensors to measure the force-time function at dif-
ferent locations [9-21.
The results from these measurements allow the following interpretation: Most
forces are transmitted at the bumper height approximately 400 to 500 mm
( 1 6 to 20 in.) above the ground. This is not valid for vehicle types other than
passenger cars. Relatively low forces are recorded from the bottom to 400 mm
( 16 in.) and above 700 mm (27.5 in.).
The type of barrier significantly influences the acceleration and deforma-

tion level of the tested vehicle. Although during the straight central impact
Figure 9.3 Measzrvernent device at thefiont of the bavvieu:

(Source: Ref 9-2.)
against a fixed wall which is due only to the vehicle asymmetry, small lateral
accelerations occur. Other crash modes clearly show visible deceleration in the
x and y directions. Also, the shape of the velocity change versus time function
is different. Figure 9.4 demonstrates the deceleration as a function of time
for a 30" barrier, a frontal 0" crash, and an offset deformable barrier crash test.
Likewise, the change in velocity is shown for the various crash types. Because
the crash against the 30" barrier was performed only at 48 kmlh (30 mph), only
the tendencies of the s, s curves as a function of time are relevant.
From the data measured, we can conclude that the deceleration level is high-
est for the frontal 0" crash. This also means the strongest requirements with
respect to the performance of the restraint systems. The offset crash produces
a higher loading on the vehicle structure because often only one front longi-
tudinal beam is taking the highest load in relation to the energy absorption.
Other frontal collision tests that are used by some vehicle manufacturers are
discussed next.
Time [s]
0 0.05 0.1 0.15 0.2 0.25

Time [s]
Figure 9.4 Deceleipation and velocity time,fi~nction,for

different
types of crashes. (Source: Re$ 1-1.)
9.1.1 Pole Test

+
With impact speeds up to 29 -t 0.5 k d h (18 0.3 mph), the test is performed
against a rigid pole that has a typical diameter of E 250 mm (E10 in.).
9.1.2 Frontal Car-to-Car Crash

For compatibility tests, see Chapter 13, for accident simulation tests, and
for developmental reasons, vehicle manufacturers also are undertaking
vehicle-to-vehicle tests. In a frontal collision between two vehicles with the
masses ml and m2, the following relationship is valid:
where
v, is the impact speed

u is the common speed after the impact
If we insert the relative velocity (v,) at the start of the impact between both
vehicles
the two preceding equations result in
With this expression, the change in velocity for Vehicle 1 and Vehicle 2 is
as follows:
and
It is interesting to see that the change in velocity is influenced only by the

vehicle masses involved and the relative collision velocity. A frequently
discussed question is: Which velocity must be chosen in a collision against
a fixed wall to simulate a frontal collision of two identical vehicles against
each other? For further evaluation, we define vi, equal to vi2, that means
that each vehicle has the same but opposite speed. One condition is that the
change in velocity AvlB must be the same as during the vehicle-to-vehicle
and vehicle-to-barrier test. For a non-elastic impact, the following relation-
ship is relevant:
Case 1: Impact Against a Fixed Barrier.
With m2 equal to the barrier with a mass and vizequal to the speed
00
of the fixed barrier, this means zero. The change in velocity in the
crash against a fixed barrier is AvlB = vil.
Case 2: Impact Against a Second Vehicle.
With m2 equal to m , (identical vehicles) and vil equal to vi2 identical

but opposite speed, the change in velocity in a vehicle-to-vehicle test
becomes Avlveh= Vil.
From this short analysis, we can see that a collision against a fixed barrier
with vil has the same consequences with respect to the change in velocity,
a non-elastic impact assumed, as a vehicle-to-vehicle collision, with each
vehicle having the same mass, structure, and impact speed. To make this
result more transparent, the following example is used: A vehicle impact
with 50 kmlh (3 1 mph) against a fixed barrier has the same consequences
as an impact of two identical cars with the same mass with a relative impact
speed of 100 kmlh (62 mph). If we consider that for the production car we
must take into account 10% elasticity, the 50 km/h (31 mph) barrier impact
has a change in velocity of 55 k m h (34 mph), and the V,,, crash of 100 kmlh
(62 mph) between two cars becomes 110 kmlh (68 mph).
The change of velocity depends on the vehicle masses. Figure 9.5 shows
the Av and the deceleration as a function of collision time for two cars with
ml = 966 kg (2 130 lb) and m2 = 1960 kg (4321 Ib) (ml/m2= 112.03). The
relative velocity between both cars at the beginning of the impact was
Velocity
16 1 I I I I 1
Time
20 40 60 80 I00 mls 120
Time
20 40 60 80 I 0 0 mls 120
Figure 9.5 Frontal collision between cars of d$erent masses.

In the relevant time period for the occupant protection, the Av for the heavier
vehicle is 10 mlsec (33 ftlsec) and for the lighter vehicle is 20.5 mlsec (67 ftl
sec). This is very close to the results of the formula mentioned previously.
1960
Av,, = 28.- =18.76 mlsec
2926
9.1.3 Design Requirements of Frontal Collisions

Regardless of the occupant restraint system, some basic requirements must be
fulfilled by the vehicle structure in frontal collisions. The longitudinal front
bars must be designed to be as horizontal as possible. In a front-wheel-drive
configuration,the driveshafts usually require an opening, which gives to the lon-
gitudinal beams a small upward tilt, followed by a downward configuration, the
so-called S-shape. Therefore, the longitudinal beams in this area are reinforced
by metal shields to avoid bending. The force from the beams is transmitted
to the outer part of the floor panel and the middle tunnel of the vehicle. Also,
the front transverse beam must be strong enough to not only resist the forces
created in low-speed impacts but to engage both front longitudinal beams in
offset crashes. The height of the longitudinal and transverse beams should be
between 400 to 500 mm ( 1 6 to 20 in.) above the ground.
Although the upper part of the front end is not the major component with
respect to energy management, it contributes to maintaining the survival space
for the occupant. This is important not only for the function of the restraint
system, but for the necessity to open one door without tools after the crash.
The sidewalls must be firmly connected with the wheel housing, and the
force in the middle of the A-pillar must be transmitted via the door and its
reinforcement (e.g., the side beam, to the B-pillar).
Another important requirement is the integrity of the windshield. The lami-

nated windshield must remain in its frame to act as a reacting part for the
airbag. In accordance with FMVSS 212 [9-31, the front hood must not
penetrate the windshield from the outside in a predefined zone. Figure 9.6
[9-41 provides a good estimate of how the energy absorption is distributed
via the front end of a vehicle in a frontal collision. If we sum these forces,
approximately 50% of the energy is absorbed by the longitudinal beams.
Second half
First half
Figure 9.6 Energy absorption o f different vehicle parts.

(Source: Ref: 9-4.)
In addition to the fact that the steering wheel requirement technically is

overruled by FMVSS 208 [7- 131, in accordance with FVMSS 204 and EEC
741297lECCR 12, the steering wheel must not have a relative movement to the
body-in-white that exceeds a horizontal rearward movement and an upward
movement of more than 127 mm (5 in.) during a 50-km/h (3 1-mph) frontal
barrier impact. The design solutions for these requirements are stable cross-
bars under the dashboard and steering columns, which have a mechanism to
collapse in the longitudinal direction during impact.
Because of the higher degree of protection, if the seat belts are used, and in
combination with airbags, other injuries become more important. This is why
special attention is given to the intrusion level ofthe foot pedals ofthe vehicle
in frontal crashes. This intrusion also is one of the assessment criteria for the
NCAP test and therefore is an important design criteria.
9.2 Lateral Collisions

According to recent accident statistics (see Figure 9.7), the lateral collision
has become a high priority. If we analyze vehicle accidents with fatalities,
the lateral impact has a very high priority.
Car occupants
killed in rear
Car occupants crashes Car occupants
killed in car-to-car killed in rollover
side crashes
Car occupants
killed in frontal
crashes
J
40%
Figure 9.7 Distribution of fatal accidents. (Source: Ref: 9-5.)
The explanation for this is the greater degree of occupant protection in

frontal impacts, which was achieved by design efforts made during the past
few decades. There are basically four legal requirements for dynamic side
impact testing:
Impact of 4000 lb = 18 14 kg rigid barrier with a speed of 32 km/h (20 mph)

under 90" to the longitudinal axis of the tested car. This test was performed
primarily to check the integrity of the he1 tank system during and after
the test and is valid only for cars built before September 1,2004.
Impact of a moving barrier in accordance with FMVSS 214 [8- 121. This test
has been the basis for legal requirements since 1993 in the United States.
The barrier with a weight of 136 1 kg (3000 Ib) is crashed with 54 k m h
(g 33.5 mph) in a crabbed configuration, as shown in Figure 9.8.
Wheelbase (W) Force
37' 0.5W
-
@ 33.5
mph TEST CONFIGURATION
53.9 kmlh
Figure 9.8 Test conjigzrration in accordance with FMVSS 214.
The impact point is defined as follows: the left corner must have a distance
of 0.51 + 0.94 m (1.7 + 0.6 ft) from the center of the rear wheel, where 1 is
the wheelbase of the impacted vehicle centerline of the moving barrier.
On the right part of Figure 9.8, we also can see the test configuration
and the force deflection characteristic of the deformable element, which
is attached to the front of the barrier. Figure 9.9 shows the velocity as a
function of time for the tested vehicle and the barrier, with the acceleration
of the impacted vehicle and of the upper spine of the dummy.
Test no. 3 is a pole test, which is specified in FMVSS 201 and is proposed
for FMVSS 2 14 (Figure 9.10). When tested for the future FMVSS 214, the
impact of the barrier is 75" to the vehicle longitudinal centerline and 90"
I6
I4
-; 12
-g I0
8
U
-0 6
P
Tlme [ms]
0 20 40 60 80 100
Tlme [ms]
Figure 9.9 Acceleration, velocity as afknction of time in a test in

accordance with FMVSS 21 4 for vehicle and occupant.
Pole to COG head Pole diameter 254 mm
I SID (50%- Male ) I c 1000
Figure 9.10 Pole test.

for FMVSS 20 I . The diameter of the pole is 254 m = 10 in. The impact
speeds are 29 kmlh for FMVSS 201 and 32 kmlh for FMVSS 2 14.
Test no. 4 is the impact of a movable/deformable barrier in accordance with

EEC regulation No. 96127lEEC [7- 151. This barrier is impacting the tested
vehicle under 90" with the centerline meeting the R-point. The details of
the barrier are specified in the amendments to the previously mentioned
rule. Of special importance is the layout of the impact device, which shows
(depending on the location of the impact element) different forceldeforma-
tion characteristics. The barrier has a weight of 950 kg (2094 Ib), and the
impact speed is 50 kmlh (3 1 mph), as shown in Figure 9.11.
Figure 9.11 Test configuration in accordance

with EG 96/27 and ECE R 95.
Figure 9.12 shows the differences of the impacting barriers as defined by the
National Highway Traffic Safety Administration (NHTSA) and Economic
Commission for Europe (ECE).
United States
Figure 9.12 DiSferences of the side impact between barrier tests

for the United States and Europe.
The ECE barrier is higher above the ground and is smaller in width than
that of the United States. Regardless of the differences in impact speed, the
deformation of the tested vehicles also differs; that is, the ECE does not hit
the body sill in most cases.
For the last two tests mentioned, it is important that the vehicle be designed
in some areas especially to resist the impacting barrier. This means a strong
A+B-pillar, strong doors and hinges, reinforcements in the doors, and a
good connection of the lower doors to the outside of the underbody with a
strong cross member in the area of the seat mounting system (front and rear).
Figure 9.13 shows the reinforcement measures for a subcompact vehicle for
side impact protection [9-61.
Figure 9.13 Crash-relatedstrt~ctziralreinforcements at the side

of the vehicle and the d o o ~
As mentioned in Table 3.1, there are several consumer impact tests performed
that use European and U.S. (FMVSS) barriers at higher speed. In the United
States, one additional side impact test is defined by the Insurance Institute
for Highway Safety (IIHS), as shown in Figure 9.14.
In general, one can state that the situation concerning worldwide requirements
is very complicated, as shown in Figure 9.15.
In the future, not only the barriers will be modified but also the types of dum-
mies used. Instead of the Euro-SID- 1, the Euro-SID-2 is used.
In the United States, the US-SID should be replaced by the EuroSID-2re,

which was described in Section 7.6 of Chapter 7.
The total complexity of the side impact regulations is analyzed under Ref. 9-8.
After the Fifteenth ESV Conference, five working groups were established to
Figure 9.14 Side impact test by IIHS [9- 71.
SW%S 101 FYH

SINCAP ( a o nm)
fm'ss 281 POM hp.d30'
TWO OOP
llHS Skh Impact (510 h)
Compatlblltty F2S ( P W u ?M)
FMVSS 211YIDB (ESZ RE)
FMISS 2 1 4 4 0 8 ($10 I6 FRO)
FHVSS 214Poh 7 P (ES2 RE)
fm'ss 2 1 4 P M 7 P (YO FRC
Eum NCAP Sld. 1mp.U (ES1)
Eum N W P P M lrnpla(ESI)
CCE R96 (EM1
ECE RS6 (ESZ)
Eum NCAP Sld. Impad (ES4
f u m NCAP Pdm hplcl(ES1)
E W C ~ n m b)(rrdlwm
r mat
-
EEVSl Japan AEYDB
Figure 9.15 Side impact ~~equirements. (New proposals

are marked by an asterisk [*I.)
harmonize the different legislation worldwide. One group is dealing with side
impact requirements. Unfortunately, the path today moves in the opposite
direction, without a positive contribution for the customer.
9.3 Rear-End Collisions

For the rear-end collision, the same test is used as for one side impact test with
the rigid movable barrier with a weight of 1800 kg (3968 lb) and a speed of
more than 48.3 kmlh (30 mph). In a phase-in time function, cars produced
after September 1, 2006 should be tested by the movable deformable barrier
with an impact speed of 80 kmlh, as defined by FMVSS 214, but in a 70%
overlap mode and with a 50-mm-lower barrier front plate. The phase-in should
be as follows: 40% of production by September 1,2006; 70% of production
by September 1,2007; and 100% of production by September 1, 2008.
In addition, the fuel tank system integrity is tested in a lateral test mode also
by the movable deformable barrier with an impact speed of 53 km/h. The fuel
tank system is checked before and after this test with the quasi-static rollover
test procedure, as described in Section 9.4.
The assessment criterion for the tested vehicle again is the integrity of the
total fuel tank system. Figure 9.16 shows the acceleration and velocity as a
function of time for the impacted vehicle.
9.4 Rollover
For the rollover situation, two different kinds of tests are performed. The
integrity of the fuel tank system is tested with the help of a test device, which
can turn the vehicle step by step up to 360" (see Figure 9.1 7).
The vehicle is mounted in this device and, via 90" steps, is tested to determine
whether the fuel tank and fuel lines show any signs of leakage. The dwell
time in each position is at least 5 min. To prevent fuel loss, gravitation valves
are installed between the ventilation pipe and the active charcoal filter.
The second procedure is used in connection with the analysis of the per-
formance of the vehicle occupants during accident simulation tests. The
Time Is]
0 0.05 0.1 0.15 03 0.25
Time [s]
Figure 9. I 6 Rear-impact data in accordance with FMVSS 301.
tested vehicle is put onto a moving sled under an angle of 23", as shown
in Figure 9.18.
The sled with the vehicle on top has a speed of 48.3 k d h (30 mph) and is
stopped via a controlled deceleration. The vehicle then starts to roll several
times, as shown in Figure 9.19
Compared with frontal and lateral impacts, the time sequence of the rollover
simulation test is much longer. Whereas the frontal and side impacts are
finished after 100 ms, the rollover is finished after several seconds-in the
case shown, at approximately 4000 ms.
Figtlre 9.17 Test conjigzlration,forfile1 tank integrity.
Figure 9.18 Test conJigzlration,for the dynamic rollover test

in accordance with FMVSS 208.
200 600 850 1350 1750 2400 2900 ms 4000

Time
Figure 9.1 9 Vehicle behavior as a,ft,nction of time during a

dynamic ieollover test. (Source: Re$ 1-1.j
The body-in-white normally has good strength to resist these impact loads.
For a two-seater vehicle, the A-pillar with the glued-in windshield and, very
often, reinforced seat backs with sufficient height offer satisfactory protection
to occupants during rollover accidents if the seat belts are worn by the occu-
pants. For some convertibles, automated roll-bars, seat-back reinforcements, or
vertical movement protection devices are used. These are activated by sensors.
A typical design for such a protection device, directly behind the rear seat
back, is shown in the 2006 Volkswagen EOS [9-91 (see Figure 9.20). Behind
both seat backs of the rear passengers, two rollover protection systems are
installed. In severe accidents (e.g., rollover or front, lateral, or rear-end col-
lisions), a pre-loaded spring is released after the sensors, and the electronic
control unit (ECU) determines the severity of the accident.
In the future, the requirements for rollover protection will increase. Not only
the roof strength in accordance with the FMVSS 216 will rise, but also the
requirements for door locks and hinges. Furthermore, a containment test in
combination with side airbag systems will be introduced for light-duty trucks
(LDTs) and SUVs and at a later date for passenger cars.
Figure 9.20 Automated rolloverprotection in a convertible.

(Source: Rej.' 9-9.)
References
Relou, J. Methoden zur Entwiclclung crashlcompatibler Fahrzeuge,
Shaker-Verlag, 2000, Aachen, ISBN 3-8265-7804-x.
Compagnie BiA. Product information, Conflans Ste. Honorine, France,

February 2002.

Mounting, National Highway Traffic Safety Administration, Washing-
Wittemann, W, lmproved Vehicle CrashworthinessDesign by Control of

the Energy AbsorptionforD@erent Collision Situations, Technical Uni-
versity of Eindhoven, The Netherlands, 1999, ISBN 90-386-0880-2.
Klanner, W. "Status Report and Future Development of the EURO

NECAP Program," Experimental Safety Vehicle (ESV) Conference
2001, Amsterdam, The Netherlands.
Heumann, G., et al. "Sicherheitskonzept des neuen Volkswagen Polo,"

Berichte des Aachener Kolloquiums Fahrzeug- und Motorentechnik,
200 1.
Insurance Institute for Highway Safety, www.hwysafety.org.
McNeil, A., et al. "Aktuelle SeitenaufprallmalJnahmen-Divergenz

oder Harmonisierung und mogliche Auswirkungen auf zukunftige
Fahrzeugkonstruktionen, VDI-Berichte, Nr. 19 1 1,2005, Dusseldorf.
Tinschert, T., et al. "Gesamtfahrzeug, Karosserie, Sicherheit. Der

EOS von Volkswagen, ATZSpecial Edition, May 2006, Vieweg-Verlag,
Wiesbaden, Germany.
Occupant Protection
10.1 Vehicle Compartment

Aside from the classic restraint system, which consists of the combination
of seat belts and airbags, the total passenger compartment must fulfill certain
requirements. For example, the minimum head impact zone is defined by a
sphere of 165 mm (6.5 in.) diameter and requires no sharp edges by a radius
of 2.5 mm (0.1 in.) or 3.2 mm (0.126 in.), depending on the location. Another
requirement is defined in the newer version of FMVSS 20 1. In this standard,
it is requested that a head form with a speed of 24 km/h (15 mph) is impacting
certain interior parts of the occupant compartment. In these impact zones,
which also include the side head airbag, special deformation measures must be
taken. Figure 10.1 shows one example of the impact test investigation using
an FEM calculation.
To keep the acceleration level low, a deformation length in the direction of

the impact of up to 50 mm (2 in.) is the basis to fulfill the specified level of
less than 80g deceleration.
10.2 Restraint Systems

With respect to restraint systems, we must differentiate between devices that
must be activated manually by the occupant (e.g., seat belts and child restraints)
and devices that work automatically (e.g., seat-belt movement limiters, belt
tensioners, and airbags). Modern seat-belt systems are extremely powerful
and provide a high degree of passenger protection.
A crucial criterion for the quality of a restraint system is the perfect combina-
tion of the vehicle structure, the steering wheel movement, the seat perfor-
mance, the occupant compartment layout, and the seat-belt characteristics.
Tesl device
A-pillar lining
- _ __ --
-HIC calculat~onfor all Impact pomts In Ihe veh~cl

oflhe necessary deformalmn behween
lining and body-in-white I
-Development of padding measures
Figure 10.1 Finite element sitzrationfor head impacts, in accordance

with FMVSS 201. (Source: Ref 10-1.)
The high performance of computer simulation procedures allows the optimal

design of occupant protection systems in cars.
10.2.1 Seat Belts

The standard design for the seating position of the front and rear occupants
is the three-point automatic belt. The inner seat-belt latch is mounted for
movable seats at the seat frame. The upper anchorage point is mounted at
the B- and C-pillars. For several years in the United States, passive seat belts
were standard equipment for the front seats.
The permissible field of the seat-belt anchorage points is defined by the

legal specifications FMVSS 210 and the EC (European Community) direc-
tive [lo-2, 10-31. In nearly every car, the upper outside anchorage point is
adjustable in the vertical direction for comfort reasons. Figure 10.2 shows
the design of such a vertical adjustment.
A feeling of comfort goes with the protection function of the seat-belt sys-
tem. This means that if the upper torso belt has a good geometrical position
Occupant Protection
Figure 10.2 Seat-belt anchorage

height adjustment.
in relation to the chest, the protection in an accident also is good. For the
locking of the seat-belt retractor, two different mechanical systems are used.
One system uses the acceleration or deceleration of the vehicle. Different
kinds of pendulums are used as locking mechanisms. As a second type of
locking device, the seat-belt pull acceleration is taken, where above a defined
value (e.g., greater than lg), the locking system must be engaged with a belt
extraction of less than 50 mm (2 in.).
For the optimal combination of the seat belts with the vehicle structure, many
subsystems exist.
The belt clamp locks the belt above the seat-belt retractor after the retractor
is locked. With this feature, the remaining belt on the retractor is prevented
from contracting more than 10 cm (4 in.), which is the length that could
be exceeded due to the relative movement of the occupant to the vehicle.
Figure 10.3 shows an example of such a design.
After the vehicle reaches or exceeds a certain deceleration level, the mechani-
cal belt tensioner is activated by a spring, which pretensioned the belt in a
time of 10 ms with a force of up to 2000 N. The mechanical belt tensioner,
as shown in Figure 10.4, often is part of the belt latch.
Figure 10.3 Seat-belt clamp mechanism.

(Source: Autoliv.)
Today, more and more pyrotechnic belt tensioners are being used. These
also have replaced designs such as the Audi PROCON-TEN system [lo-41.
Depending on the severity of the accident, the pyrotechnic belt tensioner tight-
ens the belt after exceeding a certain deceleration level. Figure 10.5 shows a
tensioner that is mounted above the seat-belt retractor in the B- or C-pillar.
Other designs have the locking and tightening parts incorporated directly
into the seat-belt retractor. In particular, the pyrotechnic seat-belt retrac-
tor allows greater freedom in the geometric design and in the layout of the
motion/time function of the occupants.
New designs have incorporated an adaptive belt-force limiter with a dual-

stage torsion bar [lo-51, as shown in Figure 10.6. With this new design, the
upper torso force level is reduced if the high force is acting for too much
time.
Occupant Protection
Figure 10.4 Mechanical seat-belt tensioneu:

(Source: Autoliv.)
Normal function Belt retracting by

belt retractor prelensioner
Figzwe 10.5 Pyrotechnic seat-belt tensione~:
155
r...7. -.. .4;
--
-
- ~p - ,-.
'
LI
.. , ,
-*
Torslon Bar L.'

.. .
~ ' 1-1 EnorgyAbsorption Ama
-
Ring
Sw~tch~ng
Switching Ring
L L
,
Figure 10.6 Principle of the adaptive beltforce limitation

with dual-stage torsion bar: (Sozrvce: Autoliv.)
Occupant Protection
For occupants to be willing to use the seat belts, comfort is of great impor-
tance. The installation of the seat-belt latch at the seat has already made a
positive contribution in that regard. Another design criteria is an optimiza-
tion between the belt retraction force and the force to the occupant's chest in
the worn condition. The minimum retraction force is necessary to guarantee
perfect retraction. The retraction is influenced strongly by the belt material,
the design of the upper belt buckle, and the belt guide above the retractor.
Humidity also influences retraction, and some advanced designs use roller
freewheels at the upper buckle to reduce beltlbuckle friction. This conse-
quently allows the reduction of the belt retraction force as well. Figure 10.7
shows an optimized design.
Figure 10.7 Example of a rzm-lhrozrgh buckle

at the upper anchorage point.
Other solutions for reducing the retraction force are on the market. For
example, a locking mechanism always locks the retraction mechanism in the
last position that the occupant has chosen.
With respect to comfort and seat-belt usage, passive seat belts were used in
the U.S. market for more that 15 years. The reason for this development was
the requirement by the National Highway Traffic Safety Administration for
passive restraint systems. The rate of seat-belt usage at that time was low in
North America, and the penalty for not wearing a seat belt was not introduced
at that time. Furthermore, because airbag systems were not advanced techni-
cally enough to be sold, passive seat-belt systems were developed. These
consisted of either passive two-point body belts with a knee bolster or passive
three-point belts. The first car on the market with the passive seat-belt system
was the Volkswagen Golf (see Figure 10.8).
Figure 10.8 Volkswagen passive seat belt. (Source: R e t 10-6.)
The body belt was attached on one side at the upper part of the door and on
the other side at the seat, where the run-through buckle and retractor were
mounted. Part of the belt also had an emergency latch and was attached via
bolts to the seat-belt anchorage points at the B-pillar. Ifthe latch was opened,
the engine could not be started. To avoid "submarining" of the occupant under
the seat belt, the seat plate was reinforced, and a knee bolster was installed in
front of the occupants. In later versions, there also were some designs where
the upper part of the seat belt was moved at the roof. If an occupant opened
the door of the vehicle, the upper part of the seat belt was pulled into a forward
position to allow easy entry and exit from the vehicle (see Figure 10.9).
Occupant Protection
Figure 10.9 Layout of an automatedpassive seat-belt system. (I) Drive

zmit; (2) Switch jor B-pillar with latch control; (3) Locking device with
emergency latch and deforination element; (4) Pansport rail; (5) Belt
latch; (6) Slip device; (7) Switch for A-pillar; (8) Warning light;
(9) Ignition; (10) Warning relay; (11) Knee bolster; (12) Switch ieearward
gear, (13) Switch for sensor locking; (14) Door contact switch;
(15) Retractor with sensor locking; (16) Acceler*ationjor driver and
passenger system. (Source: Re( 110-6.)
With respect to avoiding submarining, many alternatives were available on

the market. These ranged from knee bolsters alone to actively operated lap
belts and a lap belt integrated into the automated three-point belt. With the
lap belt, the head and neck loading often increased because of the downward
pulling force created by the vertical forces of the lap belt.
Influenced by the progress in airbag development and performance and

because more U.S. states have introduced seat-belt usage laws, the passive
seat-belt system has been replaced by a combination of three-point belts with
airbag systems.
10.2.2 Airbags
Development of the airbag system began in the late 1960s [lo-71. At that
time, the system used compressed gas as the deployment means. Although
performance was good in accident simulation tests, the influence of outside
temperatures changed the pressure in the airbag gas bottle significantly. At
high temperatures, the pressure was too high; at low temperatures, the pres-
sure was too low. In any case, airbags did not perform well. Either the bag
pressure was too high, putting the occupant at risk of injury, or the bag failed
to inflate. It took some time before a pyrotechnic solution was ready for
introduction into production cars to avoid these disadvantages.
American legislative requirements to test airbag-equipped cars with and

without the use of the front seat belt led to a complicated development of the
airbag system design. The decision by the National Highway Traffic Safety
Administration to define, via FMVSS 208, that the test should be performed
also with unbelted dummies was based on the low seat-belt usage rates at
that time, particularly in some U.S. states where no seat-belt usage laws were
enforced.
We believe that airbags are only a supplement of the main restraint system:
the worn seat belt. The public should not feel safe in any case that the vehicle
itself gives adequate protection if occupants do not wear seat belts. Mean-
while, as we will see later, many additional requirements by FMVSS 208 have
been introduced as of today, especially to cover the problem of occupants of
different sizes, seating positions, and restraint system variations (e.g., when
using child restraints). At the beginning of the rule-making period, airbags
were considered for only frontal impacts.
10.2.2.1 Airbags for Frontal Impacts

If we analyze the performance of the three-point belt alone in frontal collisions,
the occupant first moves forward relative to the vehicle interior compartment
until the belt is locked. After that, the relative movement in relation to the
body continues at a lower speed. Superimposed over this is a downward
movement of the occupant due to the effect of the lap belt. This effect
and the limitation of the belt elongation due to its design create at the final
Occupant Protection
phase of the accident a stronger head rotation. This head rotation is reduced
significantly by front airbags. Also, direct contact ofthe head against the vehicle
interior is prevented. Figure 10.10 shows the layout of an airbag system for
front occupants in frontal accidents.
Figure 10.10 Layout of an airbag system ( I ) Driver module; (2)

Passenger module; (3) Control box with sensor; (4) Failure indication
lamp; (5) Safety belt; (6) Diagnosis connection.
The system consists of several components: two pyrotechnic powder contain-

ers, the driver and passenger air bags with relevant covers, the electrical wire
harness with several sensor units, the special power transmitter for the ignition
of the steering wheel airbag, and the electronic control unit (ECU), including
the diagnosis system. For gas generation, purely pyrotechnic powder or some
hybrid designs (pyrotechniclgas) are used. The pyrotechnic material should
create the least amount of emission components as possible, although the
number of critical parts is small. The bag material consists of polyamid (PA)
or polyester (PETP) partially with blow-out holes for a controlled ride downward.
After a crash of a severity at slightly below the 20-kmlh ( 12.5-mph) barrier

impact speed, the ECU calculates in a threshold and deceleration time analysis
whether or not the ignition of the gas generator should be fired. After ignition,
the pyrotechnic material creates a gas that fills the airbag. In a 50-km/h (3 1-mph)
barrier impact, the sensor ignition time is less than 30 ms. The initiated airbag
pressure lies between 1.8 and 2.2 bar. Figure 10.1 1 shows the principle hnc-
tion of an airbag system in connection with a three-point belt.
Start of Impact Airbag Ignition Bag Deployment Occupant Contact End of Impact
Figure 10.l I Injation of driver andpassenger airbags. (Source: Rej I O-8.)
Because of the American experience with an airbag system that also must
work both with and without a seat belt, the airbag pressure had to be much
higher in the case where no seat belt was used. If the seat belt is used,
basically only the head must be prevented from rotation. Without the seat
belt, the mass that must be absorbed is more than 5 times higher. An easy
comparison demonstrates this: the reduced head mass of a 50% male is
approximately 6.8 kg (15 lb), whereas the reduced body mass is 36 kg
(79 lb). This high pressure and the inflation speed of the airbags of up to
100 kmlh (62 mph) has created several accidents with severe injuries in the
years 1995-2000. Figure 10.12 shows a distribution of lives saved versus
deaths related to airbag performance, as of 2000.
Occupant Protection
Deaths
Lives Saved
64
70
18
----.----'
[ o ~ i v e Saved
s BAddults 0 ~.- h i -
l G e m ]
Figure 10.12 Airbag safety eflectiveness.

(Sozwce: Re$ 10-9.)
New statistics show that in the meantime, because of improvements to the

vehiclelairbag system, the number of deaths caused by inflating airbags has
been reduced from a peak number of 65 in 1995 to close to zero in 2004 (see
Figure 10-13) [ 1 0- 101 and decreased further.
In most cases, the severe injuries were consequences of incorrect use of the
car seating position and seat belt. These can be categorized in the following
groups:
An out-of-position situation
Wrong type of child seats and incorrect mounting position
Children directly in front of the passenger front airbag
The immediate reaction to these accidents by vehicle manufacturers in most

European countries was to allow the front passenger airbag to be disconnected
(e.g., if child seats are installed). Meanwhile, some manufacturers offer dual-
stage airbag systems. This means that in crashes with low severity, the bag
inflation time is longer and the pressure is smaller than with high-speed
accidents. Also, the standard FMVSS 208 has been revised to a great extent,
10
adull passenger
1990 91 92 53 94 95 96 97 98 99 2000 01
model year
Figure 10.13 Deaths caused by inflating airbags, 1990-2004.

(Source: Re$ 10-10.)
and European legislation and NCAP tests strongly influenced airbag design for
frontal impacts. As a result of these changed requirements, we find that the
airbag systems will have an even more complex design that is known as smart
restraints. For frontal impacts, other types of airbags are installed in some cars,
such as the knee bag. Revised FMVSS 208 of May 2000 [lo-1 11 has many
new requirements, which are shown in Figure 10.14 [lo-12, 10- 131.
The expansion of FMVSS 208 should increase the level of protection for a
wider group of vehicle occupants and simultaneously minimize the risk for
occupants of different sizes, occupants who are belted or unbelted, out-of-
position passengers, and small adults, children, and infants.
If we compare these requirements with the rules of today introduced in the year
2000 due to the additional tests (e.g., with the 5% female Hybrid I11 dummy)
and the necessity to prove the belted and unbelted cases, the experimental
work also will increase. Furthermore, the additional definition and use of a
neck injury criteria has increased the development effort.
Test to minimize the risk of being injured by
airbags for infants, children, and
small adults
Rearward-facing child seat 3-year and 6-year

with I-year dummy
Cut off in out of
Low-risk Cut off in out of Low-risk

deployment position deployment
Low-risk
deployment
Figure 10.14 Test requirements,for FMVSS 208.

For the second group (i.e., small adults, children, and infants), the require-
ments were defined as shown in Figure 10.15.
As stated, these new requirements should mean that a wider group of vehicle
occupants will be protected. A specialty is the choice of the protection strat-
egy. For the one-year-old child dummy, we can select the airbag cutoff or the
low-risk deployment. In this case, the airbag is ignited with a performance
stage that is determined in a pre-crash:
For the one-year-old child dummy with a rearward-facing child seat with
a 64-kmih (40-mph) test against a rigid barrier
For the three- to six-year-old child dummies at the passenger seat and
for the 5% female dummy with the 26-kmih ( 1 6-mph) crash against the
rigid barrier
Other options are cutoff in out-of-position situations and cutoff if the three-
to six-year-old children are in front of the dashboard. This requires that the
vehicle must be equipped with dual-stage airbags. Table 10.1 provides a
good overview related to the complex test configuration and possibilities for
complying with FMVSS 208.
Figure 10.I6 shows one possible trigger strategy for the ignition of the airbag
system, where the center and front sensors are used [lo-131. This strategy is
related to the deployment threshold, where the obstacle, the closing speed,
the impact direction, and the impact configuration play important roles. The
legal requirements and the real-world accident situation strongly influence
the deployment of air bags and the ignition of the pretensioner of the seat-
belt system.
Although not requested by FMVSS 208, car manufacturers test other occupant
sizes, such as the 95% male. It is necessary to test not only the dual-stage
design but the optimization of the airbag itself, size and ventilation holes, and
the identification of occupants in the front seat. This means that the child
seat must be recognized in the passenger seat, where the seating position of
the passenger also must be identified, and several other techniques such as
infrared radiation or ultrasonic must be used in combination with the seat
sensor. Figure 1 0.17 shows a layout in principle.
I
Test requirementsto improve the
protection of vehicle occupants (belted
and unbelted) and of different sizes.
L-r' 59/0 Hybrid Ill

r-l
5% Hybrld Ill female
0 G' Flat barrier
I
Flat barrier
40% offset
deformable barrier
Unbelted Unbelted
3240 kmlh 32-40 kmlh
Figure 10.15 Test to minimize the risk to small adztlts,

children, and infants.
TABLE 10.1
REQUIREMENTS OF FMVSS 208
AS OF SEPTEMBER 2003
DummytTest 50th Percentile 5th Percentile 6-Year-Old 3-Year-Old 12-Month-Old
Requirement Adult Male Adult Female Child Dummy Child Dummy CRAB1 Dummy
Rigid Barrler Crash Test. Belted. X' X NIA NIA NIA
48 kmlh (30 mph);
Perpendicular
Rigid Barrler Crash Test: Belted; X NIA NIA NIA NIA
56 kmlh (35 mph),
Perpendlcular
Rigid Barr~erCrash Test, X X NIA N/A NIA
Unbelted: 32-40 kmlh (20-25 (Perpendlcular
mph), Perpendicular and 30" Only)
Offset
Offset Deformable Barrier Crash
Test (Driver's Side); Belted.
40 kmlh (25 mph)
Aulomallc Suppresston Option NIA NIA X X X
(Slalic tesllng to determine if the
airbag is autornalicaliy
deactivated when an infant is in
a car seat or a chlld is In the
passenger seal in the posltion(s)
specilied in Ihe standard)
Low-Risk Deployment Option NIA
(Vehicle must meel injury crlteria
specilicatlons when the driver or
passenger side airbag is
deployed as speclfled in the
standard)
Out-of-Position Dynamlc
Aulomalic Suppression Option
'
4 8 . ~ 111 (30.1n~I.) ,sod lo' roll ces u, I1 Illro.gl 1I.e urul.cllw y t d r pr1d.o
lor1 LS 119 male d,wu ts 1r1 911103. altef
9!1!07 tnc n gn-spccarlgd (56-kmln'3j-mpn])barf cr tcsl &. ng male a m m cs s rcq~lrca
Type A Type B Type C Type D

Left and right UFS Left and right UFS Left and right UFS CCS, left and
measure the same measure significantly measure high right UFS
acceleration different acceleration acceleration; CCS measure the
measures low same
acceleration (first acceleration
1
10-15 ms)
100% barrier
Car to car
Center pole
; g;zs;:h
Truck under-ride Curb test
AZT
Figure 10.16 Crash-type class@ation CCS end UFS sensors.

(Source: Re$ 10- 13.)
Occupant Protection
2-4 Ultrasound sensors

cov,ering the seat and
the
Figure 10.17 Ultrasonic sensors Jor occupant protection.

(Sozrrce: Re$ 10-13.)
10.2.2.2 Side Protection by Airbags

As a result of the great success of frontal airbags, side airbags in the upper
torso area were introduced first by Volvo [lo-141. Special design features for
this system are sensors for evaluation of the lateral acceleration and airbags,
mostly installed laterally on the outside of the seat back. Because the free
distance between the inner door and the occupants is much smaller in a side
impact compared to frontal collisions, all events related to airbag perfor-
mance must occur more rapidly. For example, the side impact sensor requires
a strong connection between the lower part of the door and the outer part of
the floor panel and the mounting location of the sensor. The results with the
torso airbag alone were not efficient enough with respect to head protection
in side impacts. For this reason, the head side impact airbag mounted at the
outside of the inner roof was a positive step in further reducing injuries in
side impacts. For side impact protection, because of the complex accident
mechanism, a combination of seat belts, strong seats, good design of the
occupant cell (i.e., integrity under high load), a stiff connection between the
outer parts of the vehicle and the sensor mounting place, and the installation
of a torso and head side airbag are the basis for good protection. Figure 10.18
demonstrates a head side airbag, which also protects the rear passenger, from
the A-pillar as far as the D-pillar. Nearly the entire length of the windows is
covered [ 1 0- 151.
Figttre 10.18 Side protection system in a pole impact

crash simtdation. (Source: Re$ 10-15.)
In Figure 10.19, we can see that even in a convertible, an inflatable curtain bag
is possible. The Volvo C70 folding hardtop convertible uses a door-mounted
device [lo- 1 61.
10.2.2.3 Additional Airbag Applications

In addition to front and side airbags, other airbag applications are possible,
as follows:
Occupant Protection
Figure 10.19 Door-mounted side head airbag [ I 0-161.
The side curtain airbag also could be used in rollover accidents to prevent
head contact with the vehicle interior. Because a rollover takes more time
(more than 5 sec) compared to milliseconds in frontal and side impacts,
the airbag design must be changed in a direction to keep the airbag filled
over a longer period of time.
The knee area also has been investigated to reduce possible knee and
femur injuries by the installation of a knee airbag at the driver and passen-
ger seats. One initial application can be seen in the BMW 7er (Fig. 10-20)
as of November 2001 [lo-51. Other vehicle manufacturers followed this
installation, especially in the upper class segment.
Another application is under investigation to fulfill future requirements for

pedestrian protection. In this case, an airbag system could be installed at
the front hood of the vehicle to reduce the deceleration level if the child
and adult heads are impacting the front hood.
Figzrre 10.20 Knee airbags for the driver and front

pmsenger: (Source: Reb 10-5.)
10.2.2.4 Sensors for Restraint Systems

Crash Severity. As shown in Figure 10.2 1, the number of sensors and actua-
tors has increased continuously during recent years. In a Ph.D. thesis [lo-171,
it was found that the data from all sensors (front and side) could be used to
better determine the crash severity and direction of the main impact force.
The rollover sensors provide an additional signal for this specific accident.
In all relevant accidents, we must decide whether the airbag should be fired
and at which level (Stage I or Stage 2), and whether the pretensioner of the
seat belts should be activated (see Figure 10.22).
The vertical axis of Figure 10.22 shows the crash severity and the horizontal
v-close. The v-close means the relative velocity between two colliding objects.
In general, we could say that the combination of front and side sensors with a
central sensor gives the highest degree of opportunities for correct determina-
tion of the accident severity.
Occupant Protection
3'6 Generation
1996 2"6 Generation
0 5 10 15 20 25 30
No. of Actuators
Figure 10.21 The nzrmber of sensors and actuators, as a function

ofyears. (Source: ReJ: 10-13.)
Typical Crashes
Crash Tvue Bell Airbag 1" stage Airbag 2ndstage
Rig~dbarner 0'
Angular crash
r ~ g ~ban~er
d
Pole
Otfsel crash
deformable barrier
Truck undcrride
Car-lo-car side
Misuses
Figure 10.22 Typical deployment strategy for a dual-stage airbag.

(Source: Re$ 10-13)
In side impacts, the time for the decision to fire the airbag is due to the geo-
metrical situation, which is even lower compared to that of frontal collisions.
Therefore, in addition to the g-sensing sensor, a pressure wave-based satellite
sensor is used (Figure 10.23) [lo- 181.
Intrusion d door skin during a wde Ouler door skln

collision leads to reduction d
volume within the door
This volume reduction auses a
pressure wave due lo adiabatic
pressure rise within Lhe door cavity
\
-Satellite-4
‘. ,
,,,
- -
-.
.,.A
.
Change of pressure eignal over lime
gives information about intrusion
* ,I.. 4'
. ---+ --
- - -
velocity and severity of impact
Re-enforcement bar
Very fast and reliable discrimination
of firelnowfire situations
Figure 10.23 Pressure wave side impact sensor [I 0-181.
For the reason already mentioned, pre-crash sensors also could be used in
the future, such as a radar sensor that is working up to 10 m (33 ft) ahead
and perhaps to the side. Some typical data are for the beams: 20' wide, 12'
high, wavelength 905 nm, accuracy 0.1 m (0.33 ft), velocity range 4.8 to
190 km/h (3 to 118 mph) [lo-191. This might help to identify the type of
closing obstacle. Because radar also cannot provide a determination with
respect to mass and stiffness of the crash partner, the pre-crash sensor must
be used in addition to the system already mentioned.
Sensors in the Vehicle Compartment. The interior sensing systems must

identify occupants (adults and children) on all seats. For the optimized per-
formance of the airbag system, we need much more information about the
vehicle occupants, such as why the airbag should be fired on the passenger
side if no occupant has taken the seat, or, as another example, the passenger
airbag should not be fired if a rearward-facing child seat is installed. Another
important piece of information is the weight of the occupant, his or her size,
Occupant Protection
and the seating position, particularly at the moment of the crash. Two types of
sensor systems will be used in the future. One measures the pressure and the
pressure distribution at the seat with a flat seat mat (see Figure 10.24),which
can identify to some extent whether humans or child seats are occupying the
relevant seat or whether the seat is empty.
Figure 10.24 Child-seat recognition system on the passenger seat.

(I) Resonator # I ; (2) Resonator #2; (3) Electronic device for child-seat
detection and passenger presence detection; (4) Antenna A; (5) Antenna B;
(6) Passenger presence detectov: (Source: ReJ II-4.)
The other type of sensor is used to determine the initial occupant position,
because it is important to avoid situations in which the airbag could create
negative effects. In this case, ultrasound or infrared sensors are the technical
solutions.
Figure 10.25 shows a Photonic Mixing Device (PMD) that also can provide
a determination about occupant movement [lo-201. The device has the fol-
lowing features:
PN-Modulated
Light Emitter
PMD Array mxn Times

(per Pixel)
Figure 10.25 Photonic Mixing Device. (Source: Ref 10-20.)
Emission of "RF" modulated incoherent light

Reflection of desired three-dimensional object
Detection of reflected light
Mixing with originally emitted signal on chip
Gray-scale image
Distance information from each pixel
These sensors could suppress the airbag activation if occupants are in the non-
firing zone. A typical situation is a child standing in front of the dashboard or
if the occupant is out of position. Figure 10.26 shows a typical firing decision
matrix for driver and passenger [lo- 131.
Identification of Child Seats. [n cases where the ISOFIX child seat is used,
sensors in the two locking mechanisms also could identify a mounted ISOFIX
child seat at the front passenger seat, although today no production solution
exists. In this case, the airbag in front of the child seat is not activated.
Occupant Protection
Firing-Decision Matrix:
Driver Passenger
belted I weight I position Ifolw. displ . 1 side displ .
ves 1 class 3 Ibackward 1 6 cm 1 2cm
1 Rear Passenger 1 11
&II Rear Passenger 2
Rear Passenger 3 11
Figure 10.26 Firing-decision matrix. (Source: Ref 10.13.)
Figure 10.27 shows the large number of sensors and actuators for the different
fields of application [ I 0-2 I]:
Accident avoidance
Mitigation of injuries
Driver assistance
Autonomous driving
From Figure 10.27, we also can see that some sensor fusion seems possible
in the future. It is important that driver assistance systems are developed
and installed, which both support the driver and passenger and are accepted
by them.
PRE-SAFE Systems. An advanced protection system was developed by

DaimlerChrysler and is called PRE-SAFE [I 0-221. As shown in Figure 10.28,
the system increases occupant safety after the system has recognized that a
severe accident might occur.
Accident Avo~dance Semi-Autonomous Driving
Active
Active Safety Vehicle Guidance
Collision
I Parking brake I Lane keeping
avoidance ass~stant
Predlct~veactwe
77 GHz
1
I
Pedestrian :partner
oroledion
1
I 1
{'' Video Nav~ ,
Safety I Vehicle I Comfort

panoramic view us
Pedestrian I object
recognition (Rear
24 GHz
view) , assistant
Pre-crash
sensor
1 Accident alert 1
Passive Safety Driver Support
Passive
Figure 10.27 Sensor applications. (Sowee: Re$ 10-21.)

Occupant Protection
b Seat-belt tensioner pre-activated
b Seat back inclined, headrest put up

into the most upward position
b Sunroof closed
b Knee protector, if available, locked

and side door panels put into position
b Automated pre-accident braking
Figwe 10.28 PRE-SAFE system.

(Source: Ref 10-22)
Several protection items can be pre-activated. For example, there is the seat
belt tensioner, the seat back is tilted into an upright position, the headrest is
brought into its position, the sunroof could be closed, and vehicle parts such
as armrests could be brought closer to the occupant. Although not all of
these items might be seen in future production cars, the research approach is
interesting. Some of these ideas are already in production.
10.3 Child Restraints

In accordance with legal requirements (e.g., in Germany), children who are
younger than 12 years old and are smaller than 150 cm (59 in.) in height
must be placed in special child restraints in vehicles. Child restraints must be
designed in conformance with the regulation ECE-R44 [lo-231. Figure 10.29
shows an overview related to different child restraint systems as a function of
the weight of the child [lo-231. The performance of the child restraint system
is based on the limits defined in Table 10.2.
Group 0: up to 10 kg Group 010+: up to 10113 kg Group 011: up to 18 kg Group 1: 9 - 18 kg
Baby carrier I Baby seat I Reboard seat 5-point belt system
Group 1: 9 - 18 kg I Group 1111: 9 - 25 kg I Group II: 15 - 25 kg Group 111111: 15 - 36 kg
Impact absorber I 3-point belt system I Impact absorber Booster seat withlwithout
sleeping support
Figure 10.29 Child restraint systems according to ECE-R44.

Occupant Protection
TABLE 10.2
REGULATIONS FOR CHILD SEATS
Limits for Children
Limits for Children CMVSS 21 3 [I0-261,
ECE-R44 FMVSS [I 0-271
Chest acceleration 55 g Head injury criteria 1000
(ares 3 ms) (HIC36 ms)
Chest acceleration
vertical (a, 3 ms)
Head displacement
30g
550 mm
I Chest acceleration
(ares 3 ms)
Biomechanical values
60 g
horizontal (experiences):
Neck moment (My) 20 Nm
Head displacement 800 mm Neck force (F,) 2.0 kN
vertical
Excursion limits:
I Head 8131720 mm
I Knee 915 mm
Special attention must be given to the correct use of child restraint systems. In
competitive tests, it was shown, for example, that with a belt slack of approxi-
mately 75 mm (3 in.), the child dummy measurement data were approximately
30 to 40% higher, compared to a correctly used seat belt. Two special child
restraints should be mentioned: (1) the baby carrier, and (2) the reboard seat.
In most cases, both are placed on the front passenger seat. This automati-
cally creates the previously mentioned conflict with the inflation of the front
passenger airbag. The other system is the ISOFIX child seat [lo-241, which
includes in accordance with I S 0 [lo-251 a standardized connection between
the child seat and the vehicle interior. Beginning in 2006, the ISOFIX is
allowed to be installed only with a top tether. A bracket as a third anchorage
point gives not only a better (more secure) connection but also a safer feeling
because of little lateral movement during driving. The anchorage point also
must fulfill the requirements set by ECE R14, FMVSS 225, and CMVSS 2 10.
Figure 10.30 shows the layout of an ISOFIX seat.
r
Body
1 2 3
Figure 10.30 Schematic illustration of a top-tether ISOFIX system.

1 0.4 References
Oehlschlaeger, H., el al. "FEM-Crashsimulation: Ein modernes
Werkzeug in der Nutzfahrzeug-Entwicklung," Proceedings of the
Verband der Automobilindustrie (Association of the [German]
Automobile Industry) (VDA) Technical Congress, IAA Internatio-
nale Automobil-Ausstellung [International Automobile Exhibition],
Frankfurt, Germany, 2000.
National Highway Traffic Safety Administration (NHTSA) and Fed-

eral Motor Vehicle Safety Standard (FMVSS), FMVSS 2 10, Seat belt
assembly anchorages, National Highway Traffic Safety Administra-
tion, Washington, DC, United States.
EC Directive, Brussels, 7611 151ECC "Seat belt anchorages."
Vollmer, E. Audi AG, patent description EP 0 1894 10 and EP 0 1 9 176 1.
"Der Neue BMW 7er," ATZ/MTZ, Special Publication, November

200 1, Vieweg-Verlag, Wiesbaden, Germany.
Seiffert, U. "Volkswagen Passive Occupant Protection Progress

Report-1979," SAE Paper No. 790326, Society of Autmotive
Occupant Protection
Seiffert, U., et al. "Development Problems with Inflatable Restraints

in Small Passenger Vehicles," SAE Paper No. 720409, Society of
Automotive Engineers, Warrendale, PA, United States.
Seiffert, U . Fahrzeugsicherheit, VDI-Verlag, Dusseldorf, 1992.
Author's interpretation of data in information from the National

Highway Traffic Safety Administration (NHTSA) and Insurance
Institute of Highway Safety (IIHS).
Status Report, Vol. 40, No. 7, Insurance Institute for Highway Safety,
April 6, 2005, Arlington, VA, United States.
National Highway Traffic Safety Administration (NHTSA) and

Federal Motor Vehicle Safety Standard (FMVSS), FMVSS 208,
Occupant protection, May 2000, National Highway Traffic Safety
Administration, Washington, DC, United States.
Szilagyi, I., et al. "Consequences of the New NHTSA Regulations,"

Fifth International Symposium on Airbags, Fraunhofer-lnstitut, Karl-
sruhe, Germany, December 2000, ISSN 0722-4087.
Rolleke, M., et 01. "Smart Sensors for Passenger Safety Systems,"

Fifth International Symposium on Airbags, Karlsruhe, Germany,
December 2000, ISSN 0722-4087.
Volvo company, press release.
Scheef, I., et al. "Design of Side-Protection Features-Example Audi

A4," Fifth International Symposium on Airbags, Fraunhofer-Institut,
Karlsruhe, Germany, December 2000, ISSN 0722-4087.
"Volvo Bags," Automotive Engineering Inter-national, July 2006,

SAE International, Warrendale, PA, United States.
Hubler, R. Unterstiitzung bei der Auslegung von Airbagsysternen

durch FEM-Berechnung, Dissertation, Vol. 11, ed. by Institut fur
Elektrische Messtechnik und Grundlagen der Elektrotechnik, Tech-

nical University of Braunschweig, Germany, 2000, ISBN 3-89653-
828-4.
10-18. Technical information by Siemens VDO Automotive, Germany.
10-19. Continental Temic Co. Product information material, 2003.
10-20. Schwarte, R., et al. "New Powerful Sensory Tool in Automotive

Safety Systems Based on PMD Technology," Advanced Microsystems
for Automotive Applications 2000, April 2000, Berlin, Germany.
10-21. Knoll, P, "Surround Sensing and Sensor Data Fusion," Verband

der Automobilindustrie (Association of the [German] Automobile
Industry) (VDA) Technical Congress, March 20-21,2002, Stuttgart,
Germany.
10-22. DaimlerChrysler, press release, IAA Frankfurt, 200 1
10-23. ECE-R44. "Einheitliche Bedingungen fur die Genehmigung der

Ruckhalteeinrichtungen fur Kinder in Kraftfahrzeugen," Economic
Commission for Europe (ECE).
10-24. Langwieder, K., el at. "ISOFIX-Possibilities and Problems of a

New Concept of Child Restraint Systems," in VDI-Berichte 1637,
Vehicle Concepts for the 2nd Century of Automotive Technology,
ed. by Verein Duetscher Ingenieure, VDI-Verlag, Diisseldorf, 2001.
10-25. Weber, S., et crl. Schutz fur die Kleinsten-Entwickliingspotentiale der

Kindersichesheit VDI-Bericht 1911, Diisseldorf, 2005, ISBN 3-3 18-
091911-6.
10-26. Canadian Motor Vehicle Safety Standard (CMVSS) 2 13.

Occupant Protection
10-27. National Highway Traffic Safety Administration (NHTSA) and Fed-

eral Motor Vehicle Safety Standard (FMVSS), FMVSS 213, Child
restraint systems, Code of Federal Regulations, National Highway
Traffic Safety Administration, Washington, DC, United States.
10-28. Britax-Roemer Kindersicherheit, www.britax-roemer.de.

Interrelationships Among
Occupant, Restraint System,
and Vehicle in Accidents
Although the single elements of the total vehicle, the vehicle interior, and
the restraint system have been discussed in detail in Chapters 9 and 10,
Chapter 11 describes the interrelationships among the occupant, the vehicle,
and the restraint system.
1 1 . 1 Frontal Impacts
1 1 . 1 1 The Unrestrained Occupant

In a frontal impact against a fixed barrier or another vehicle, the occupant is
moving toward the dashboard or the steering wheel due to the impact vector
created by the impacting vehicle. In the first approximation, the following
interrelationship is relevant.
Vehicle Occupant Until Impact
Vehicle Occupant Until Impact

The relative movement As between the occupant and the vehicle can be cal-
culated from the preceding formula as follows:
At a 50-krnlh (3 1.15-mph) impact with an average vehicle deceleration of

15g and a distance between the occupant and the steering wheel of 30 cm
(12 in.), the occupant hits the steering wheel after approximately 64 ms.
Figure 11.1 shows the distance between the chest to the steering wheel in
certain vehicles. From this figure, we could conclude that the 30 cm (12 in.)
distance is not unrealistic.
The differential speed between the occupant and the steering wheel remains
at 33.4 kmlh (20.8 mph) at the time the chest hits the steering wheel because
of the free forward movement of the occupant. In a more severe frontal
accident without any restraint system, the kinetic energy of the occupant
must be absorbed by the interior of the vehicle, (e.g., the steering wheel, the
- o s m +E U 9 5
E
ze
z ?=g
ex
B-
g
in F bq
6 g:
md ,E* z"
F, sg s
5~ 5 -2
> g 7 -a
u
g $a
Y-'
0 LL
LL
I
g ,,o
Vehicle
Figure II.I Avemge chest-to-steering-wheel distance, by vehicle.

{Sozwce: Re$ I I - I .)
Interrelationships
dashboard, the knee impact area, and the windshield). If we assume that for
this deceleration, a deformation length of 0.1 m (0.33 ft) is available, then
the average deceleration of the occupant is close to 50g via a time period of
approximately 25.5 ms. Figure 11.2 demonstrates this theoretical pattern.
This deceleration level of the occupant body is already three times as high as
the average vehicle deceleration level. This clearly explains two matters. If
the occupant is hitting the vehicle interior, it is necessary to prevent injuries
and, more importantly, to connect the vehicle occupant to a stable occupant
cell as quickly as possible, to allow a ride downward in the restrained system
together with the vehicle. If, for example, an unrestrained occupant is hit-
ting the vehicle interior, the impact zone must be built from energy-absorbing
material. Figure 11.3 demonstrates one technical solution related to the new
requirements of FMVSS 20 1. Of course, other energy absorbing materials,
such as polyurethane foam, are possible.
The most important element with respect to minimizing injuries is the use of
seat belts and/or the child restraint system.
1 1.1.2 The Three-Point Belt

With a three-point-automatic belt, without a belt clamping device or a preten-
sioner, the occupant is moving forward in the direction of the impact vector
of the car that is being hit until the seat-belt retractor is locked either by the
belt extension speed or by the force of gravity and until the belt is contracted
at the belt retractor spool. Through the lap belt and the upper torso belt, the
occupant's body is decelerated in its relative movement to the vehicle. With
increasing load and the forward movement of the occupant as a result of its
geometry, the lap belt creates a vertical load in a downward direction. For this
reason, the seat and the seat pan must be integrated into the optimization of the
performance of the overall occupant restraint system. Figure 1 1.4 shows, as a
function oftime, some typical data for the resulting head and chest acceleration
and shoulder and pelvic belt force for a 50% male dummy that was restrained
by a three-point-belt in a 50-km/h (3 1-mph) frontal barrier crash. As illus-
trated in Figure 11.4, it takes more than 30 ms until the belt is locked, any
slack between the seat belt and the occupant is removed, and the first sign of
an increase of the restraint force can be identified.
Tlme (msec)
t l o 2 u ~ m s Q % n ~ m m
I
Time (msec)
Figure 11.2 Deceleration, velocity, deforination, and movement,

as a function of time.
Interrelationships
Figure 11.3 Thernzoplastic honeycomb for the head impacz area.

(Source: Re$ 11-2.)
0 20 40 60 80 100 120 140 160 180

Time (msec)
Figure 11.4 Decelerations and belt forces, as a function oftime,

for a dummy restraint by a three-point belt.
For the performance of the restraint system, it is important to minimize this

time as much as possible; therefore, we must prevent the pulling out of too
much of the remaining belt from the retractor spool. The technical features
used are belt clamping devices and pretensioners. In addition, the belt slack
around the occupant must be reduced, especially when the occupant is wearing
many pieces of clothing or heavy clothing (i.e., in winter time). With the
use of a pretensioner, this positive effect is achieved. After a specific acci-
dent severity level is reached, mechanical or pyrotechnic belt tensioners are
activated. With the pretensioner, we also find some seat-belt load limiters to
avoid the shoulder belt force being too high. In the PRE-SAFE system, this
pre-loading is done by electrical motors so that the belt can be released if no
accident occurs or if the accident is not severe. Some of the systems used
are described as follows:
1 1.1.2.1 The Seat-Belt Clamping Device Above the

Retractor
This part can be used to stop a contraction ofthe remaining seat belt at the belt
spool. By a kinematic activation through a relative movement of the occupant
to the vehicle, the belt is grabbed directly above the seat-belt retractor.
1 1.1.2.2 The Mechanical Pretensioner

Basically, two systems were put into production: (1) a pretensioned spring
with a mechanical or a pyrotechnic release, and (2) as a specialty, a system
called PROCON-TEN, which was used by Audi AG for several years and is
shown in Figure 11.5 [l l-31. This system used the relative movement of the
transmission housing and the vehicle body to pull the steering wheel away
from the driver, and tensioned the belts for the driver and right front passenger,
if the accident was severe. The belt forces in the shoulder belts were limited
by force limiters to avoid chest injuries.
Most of the devices mentioned here are no longer in use today because the
pyrotechnic pretensioner is more flexible in its vehicle applications and better
in its performance.
1 1.1.2.3 The Pyrotechnic Pretensioner

Again, several different designs are in production. Most often, a pyrotech-
nic pretensioner is used either as part of the seat belt, above the seat-belt
retractor, or integrated into the retractor itself. Figure 10.5 shows a typical
layout for a pyrotechnic pretensioner.
Interrelationships
Figure 11.5 The PROCON-TEN system by Audi AG. (Sowce: Ref: 11-3.)
The pretensioning function is activated by a sensor-triggering signal. This sen-

sor could be a mechanical or an electronic sensor, together with the relevant
electronic control unit. With this technical feature, the designer has more
capability to improve the performance of the restraint system.
Figure 11.6 shows another design [I 1-41, This design of the pretensioning
device has been integrated into the seat-belt retractor and has a function similar
to the rotor of the Wankel engine. In a severe crash, the spool of the retrac-
tor is rotated via a pyrotechnic gas in the direction opposite to the pull-out
function, if the sealt belt is worn by the occupant.
Figure 11.6 Pyrotechnic rotational seat-belt pretensioner:

(Souire: Re$ 11-4.)
1 1.1.2.4 The Seat-Belt Load Limiter

To prevent too high a seat belt load, especially at the shoulder belt, various
three-point belts use an integrated seat-belt force limiter. Figure 11.7 shows
a relatively simple layout [ll-41.
By the selecting the proper seat-belt retractor dimensions and the rotating
shaft material, the force level can be predetermined. The most sophisticated
system, as of today, is installed in the BMW 7 series. The principle for this
system has already been shown in Figure 10.6. Figure 11.8 shows the load
level as a function of time [I 1-51.
The adaptive belt force limiter uses a two-stage torsion bar installed via the
shoulder belt part in the seat-belt retractor. With a pyrotechnic device, the
system can switch from a high force level to a low force level. In the initial
stage, the high force level is applied. If the accident is too severe, the system
switches to a lower level during the loading phase, based on the force applied
by the occupant to the seat belt. This allows a force time function where the
force is reduced, if the loading phase-applied through the occupant to the
upper torso belt-takes too much time.
Interrelationships
Figure 11.7 Seat-belt force limiter.
Validation Sled Test

All Load Limiter Curves
5% and 50% Dummy Types
Figure 11.8 Adaptive belf,forcelimitation with a two-stage torsion bar:

1 1 . 1 - 3 Passive Restraints
Because of the low seat-belt usage rate, especially in the United States in the
late 1 9 6 0 ~the~ introduction of passive restraints into production cars was
discussed. One solution at that time was the installation of airbags for the
front passenger. At that time, the filling process for the airbag was done by
compressed gas. Unfortunately, the pressure in the gas bottle was too sensi-
tive to temperature changes. This means that at high temperatures, the bag
pressure was too high. Likewise, at cold temperatures, the bag did not inflate
sufficiently. Even in later years, especially in the United States because there
were no special campaigns or enforcement actions, the seat-belt usage rate
was not high enough. As a consequence, the request for the introduction of
passive restraints increased again, and the development of pyrotechnic infla-
tor systems was successful.
Figure 11.9 shows for some individual U.S. states an increase of approximately
10% by buckle-up campaigns, with a usage level between 49 and 84%. There-
fore, the pressure for passive restraints never stopped. Because of the problems
Alabama Florida Georgia Kentucky Mlsslsslppl North South Tennessee

Caroilna Carolina
O Before Campaign Height of Campaign
Figure 11.9 Seat-belt ztsage rates in vaviozts states in the United States.
(Source: Re$ 11-6.)
Interrelationships
with airbags as mentioned, passive seat belts were installed in various vehicles
in the United States and partially in Europe as an interim solution.
The Volkswagen Passive Restraint System 111-7, 11-81. The Volkswagen

Restraint Automatic System (VW-RA) was the first such system in the U.S.
market. Passive seat-belt systems were either upper torso belts with a knee
bolster andlor, at a later stage, an additional lap belt or passive automatic
three-point-belts. Figure 10.8 shows the design of the VW-RA.
The upper torso belt was connected on one side via an emergency latch to
the door frame and on the other side at the seat structure. In the event of
an accident, two bolts transmitted the belt pulling forces to a plate that was
mounted at the B-pillar, where normally the upper seat belt anchorage point is
installed. In front of the front occupants, a special knee bolster prevented the
"submarining" of the occupant during frontal accidents. The anti-submarining
effect also was supported by the seat layout, where particularly the seat pan
was designed strong enough to prevent submarining. Figure 11.10 shows a
comparison of the head and pelvic motions for a three-point-belt compared
to the VW-RA.
Figure 11.10 Difference o f occtlpant kinematics for a passive body belt

and conventional three-poinf belt. (Source: Ref I - I .)
As shown in Figure 1 1.10, we could observe significantly less head rotation

and vertical ride downward for the occupant restrained by the VW-RA. The
VW-RA was offered in a subcompact car. For compact and high-class cars,
passive automatic three-point belts were installed. If a lap belt was used, the
performance was more similar to the behavior of a three-point belt.
The upper torso belt did move automatically into a forward position if the front
doors of the vehicle were opened, and it returned to position after the doors
were closed. All other elements such as the pretensioner and force limiter,
as well as an emergency locking mechanism, were possible.
Although the degree of comfort was much higher than with the standard three-
point belt, the further successful development of the airbag system caused a
change in restraint system technology.
1 1.1.4 The Airbag System

Since first introduced in production vehicles during the second half of the
1980s, the airbag system has undergone continuous development that is
ongoing today. Figure 10.10 shows the layout of a basic airbag system.
The sensor system identifies the crash severity and, for some vehicles, the
occupant seating position, as explained in Chapter 10. For the driver, the airbag
is installed in the center of the steering wheel. For the passenger, the airbag
is installed in the upper part of the dashboard. Because of U.S. requirements
for the fulfillment of FMVSS 208, the airbag system also must be tested in
the relevant vehicle without seat belts, and the lower part of the dashboard
is designed in such a way that in the event of a leg impact, no sharp edges
nor rigid vehicle parts create injuries. Furthermore, some energy absorption
to avoid too high of an upper load must be possible. After the evaluation of
the sensor signal through the electronic control unit, and if the decision is
made that the airbag should be deployed, the pyrotechnic material in the gas
generator is ignited, and the gas fills the driver and front passenger airbags.
The size of the driver airbag is approximately 60 1 (1 6 gal), and the passen-
ger bag is approximately 120 1 (32 gal). In most cases, the bag material is
normally polyamid with some pop-out holes to allow a better ride downward
of the occupant. The geometric layout depends greatly on the vehicle itself.
On the driver side, we frequently find a cylinder shape for the airbag. On the
Interrelationships
passenger side, the airbag resembles the form of a tube. Inside the airbag in
some designs are strips to prevent the airbag from becoming an undefined
shape. Especially on the passenger side, the airbag configuration also
must take into consideration the performance in an accident with an out-
of-position occupant.
In a single-stage system, the bag can reach a relative high speed of deploy-
ment of more than 100 kmlh (62 mph), when it comes close to the occupant.
In a dual-stage mode, the airbag is filled with lower pressure if the accident
is not as severe or if the occupant is located in an out-of-position situation.
Examples are the standing child or ifthe occupant is too close to the dashboard.
Also, different occupant sizes, such as a small driver sitting too close to the
steering wheel, influence the strategy of the airbag inflation characteristics
and the geometric design.
With respect to the pyrotechnic material, improvements were made to reduce

particles and carbon monoxide. New designs also are using a hybrid genera-
tor as a good compromise between the inert gas and the propellant material,
as shown in Figure 1 1.11.
Inert Gas
(=220 bar)
Figure 11.11 Hybrid gas generator. (Sozwce: Re$ 11-4.)
199
To reduce the knee and upper leg forces and the local pressure at the knee,
the first production cars also featured a knee bag installed below the steering
wheel area.
As mentioned, the airbag system is only complementary to the major restraint

system, which is the seat belt. The optimal protection function can be achieved
only via a good combination of the crash characteristics of the vehicle struc-
ture, the integrity of the passenger compartment, the seat, the seat-belt system,
and the airbag.
1 1.1.5 Comparison of Test Results for D fferent

Restraint Systems
In Figure 11.12, dummy measurement data for two different restraint systems
in a frontal collision are compared. In one case, we can see a three-point belt;
the other case shows a combination of a three-point belt with a driver airbag.
One could identify the much lower resultant head acceleration.
1 1.2 Lateral Collisions

The possibilities for applying special restraint systems for the protection of
vehicle occupants in lateral collisions are much smaller compared to those for
frontal collisions. This is due to the proximity of the occupant to the vehicle
interior side (door). One of the influencing parameters is the stiffness of the
body-in-white. The side of the impacted vehicle must be strong to show at
least some resistance against the intrusion of an impacting car. This requires
a corresponding design of the body sills, the side beams in the door, and
other door reinforcements (e.g., a connection via a hook between the lower
part of the door and the body-in-white sill). Also, cross-bars (i.e., below the
dashboard) in the vehicle and the seat design in the seat-mounting area have
a positive effect on crash protection with respect to resistance against lateral
forces, as well as cross-bars in the rear of the vehicle. The other parameter is
the layout of the side interior of the vehicle. Special attention must be given
to the size, shape, and material characteristics of the side door panel and the
head impact zone.
Intewelationships
Deceleration (a) Force (f)
Time
Deceleration (a) Force (f)
0 20 40 60 80 100 120 140 ms 180

Time
p 11.12 Dummy data. Upperediagiwnz: three-point bell. Lower

diagr#arn: three-point belt with airbag (driver.).
The very positive element of passenger protection in lateral collisions is the

installation of side airbag systems in the chest and especially in the head
impact areas.
1 1.2.1 Theoretical Analysis for Lateral Collisions

Compared to frontal impacts, the variables of an accident pattern are much
higher in lateral collisions, such as the direction of impact, point of impact,
and geometry of the obstacle. The main parameters are as follows:
Collision partner involved (mass, deformation forces, structural integrity,

and structural geometry)
Occupant size
Impact point and angle
Impact velocities
Seating positions of the occupants
Interior design of the vehicle
Use of the seat-belt system by occupants
Installation and type of side airbags
For a better understanding of this complex accident situation, Figure 11.13

explains in principle the sequence of a typical lateral collision between two
vehicles. With a simplified representation of the impacted and impacting
vehicles and the one simulated occupant as one mass, the velocity-time func-
tion of the two vehicles and the occupant is shown.
In Figure 1 1.13, the numbers I to 6 identify the velocity-time function related

to the occupant and parts of the hidden vehicle. The areas surrounded by these
velocity traces correspond with the relative distance between the identified
vehicle parts and between the vehicle components and occupants. The free
deformation characteristics of the structure and the lateral restraint system-in
Interrelationships
Ff
P
Vehicle 2
Velocity 1-C Defonnatlon vehlcle 2 = S,

1-1r Deformation doorvehicle 1
Deformation door paddlng ls~dsalrbag = S.
Vehicle 2
1- Ileformatlon pasamgsr
6'
t~ f~ f~ Time
t~ tges
Figure 11.13 Theoretical velocity-time functions for the vehicle strzrctzrre

and the occupant in side impacts. (Source: Re$ 1-1.)
this case, a cushion-have been defined to react against external forces with a
rectangular force deflection characteristic. Also, the oscillation of the dummies
or vehicle parts is not taken into consideration in this simulation. Therefore,
the results for the velocity v, the deformations, and the relative movement
of the simulated occupant as a function of time f(t) differ from the behavior
in a real accident. However, the main interaction of the occupant and the
vehicle can be understood. The impacting vehicle drives with the velocity
under 90" into the side of the impacted vehicle. 1n this case, the impacted
vehicle has zero velocity. After a short time, the outside of door 5 reaches
the same velocity as the bumper of the impacting vehicle 6. The inside of
door 3 reaches the same velocity as the bumper. The contact between the
occupant and restraint system 3 starts at the time when the occupant hits the
inner door. At the time tT, the relative velocity between the occupant and
door reaches its maximum. The relative speed could become higher as the
total change of velocity of the impacted vehicle. The occupant is deceler-
ated by the inside components of the vehicle. The deformation of the door,
including the armrest, is sR. The occupant and the bumper have at time tR
the same velocity. Until point tp,,, the velocity of the impacted occupant
remains high until the occupant comes to a standstill or hits the front-seat
passenger. The deformation of the side structure is finished at the point tF
because the seats and the underbody of the vehicle, including the cross-bars,
have no further deformation capability. The front structure of the impacting
vehicle has deformed itself by the amount of sF. The deformation of the side
structure is described through the relative movement of door 5 related to an
undeformed point of the impacting vehicle 7. Figure 11.14 shows that even
without airbags, a strong relationship is given by different layouts of the
structure and the vehicle interior, especially the door.
The upper part of Figure 1I . 14 shows a design layout with arelatively low stiff-
ness of the side structure combined with some thin upholstery of the impacted
vehicle. This means that a relatively large free distance exists between the
occupant and the upholstery. For the occupant, high loadings are expected
if the distance between the occupant and the inside of the vehicle increases.
The upholstery is deformed at point tA. The occupant then is spontaneously
accelerated to the speed of the front part 6 of the impacting vehicle. Because
of the low energy-absorption capability of the upholstery, this impacting speed
cannot be reduced.
Interrelationships
q-=-- t=
0 1
I
""""'"""=- -
2 3 4
-
lr-5 6 7
Velocity c:==:J Deformation whlcle 2 a 5t;

c:==:J Defonnatlon door vehicle 1
Vehicle 2
c::::::3] DeformaHon door padding /side alrbag " SR
c:::J Defonnatlon passenger
;·
/
1,2
4---
3,~-
Vehicle 1
tA tT tR Time
tF tges
Velocity c:==:J Deformation vehicle 2,. SF

c:==:J Delonn.tlon door vehicle 1
Vehicle 2
c::z:::::J DeformaHon door padding /5lde alrbag = SR
c:==:J Defonnatlon passenger
6/
5,
3,~_
Vehicle 1
tA tT tR Time
lF iges
Figure 11.14 Theoretical velocity-time jimctions for the vehicle structure

and occupant, with vehicle variations in side impacts.
205
The lower part of Figure 11.14 shows the other extreme. In this case, the
side structure of the impacted vehicle is stiff, and the upholstery at the door is
relatively large. It also could be a torso side airbag. Loadings for the occupant
are much less because of the energy-absorbing capability of the upholstery or
the torso side airbag and the lower relative velocity of the inner door. This
is because the contact between the occupant and the inner part of the door
occurs in a much shorter time.
11.2.2 Side Airbag Systems
1 1.2.2.1 Thorax Side Airbags

The first thorax side airbag was installed by the Swedish car manufacturer,
Volvo [l l-41. Figure 1 1.15 provides a good overview of the system. This
system consists of an airbag installed in the outer side of the seat back, a
pyrotechnic inflator, and a sensor with the control unit.
lnitial results in comparative testing showed in some cars similar deceleration

levels at the chest area, with and without installed thorax airbags. This is eas-
ily understandable if we examine the distance between the inner door panel
and the occupant. Even with a thorax airbag, the free deformation length in
the transverse direction is rather small. On the other hand, if we compare
the specific pressure at the occupant or at the dummy due to the much larger
impact area typically offered by the thorax airbag, there is an advantage for
the airbag system. This means the deceleration level is not reduced, but the
specific pressure is much lower. Figure 11.16 shows the results of a com-
puter simulation for the analysis of the protection level for a child in a lateral
impact, with and without an airbag [ll-91. The improvement is significant,
particularly with the optimized version. The two systems in production today
are torso bags, mounted either in the outer side ofthe seat back or in the inner
part of the door.
The thorax airbag also allows earlier contact by the impacting door panel.
The acceleration distance for the impacted occupant will be extended through
this effect. The seat-mounted version has the advantage, in that the position
of the thorax airbag is independent from the seat longitudinal position. This
provides not only a constant position to the occupant but allows more rapid
Interrelationships
Compression Housing Percussion cap Pyrotechnic

washer hoses
Sensor unit
Figure 11.15 Vehicle seat with SIPS bag module.

inflation because of the smaller size of the airbag. Figure 11.17 shows the
time sequence for airbag performance in a lateral collision [I 1-10].
The side airbag, including the sensor control that calculates the level of
severity of the accident, is inflated after 15 ms. The occupant is hit after 20
Protection Potential
SINCAP. Hlll 3y
Without airbag Basic -Low Risk
-
-1
HIC Nij
I
Figure 11.16 Compavison of the protection level,fov a child dummy in a
lateral impact, with and without an airbag. (Source: Ref: 11-9.)
Figure 11.17 Typical side airbag,firnction versus time.

(Sozrvce: Ref: 11-I 0.)
to 25 ms. In a side impact, the occupant can be supported by the padding
design in the pelvic area, as well as through the airbag in the chest area. Some
other solutions are the use of a special shape of side airbag, the thorax-pelvic
airbag. The decision regarding which type of airbag will be used depends on
the strategy of the vehicle manufacturer. Figure I I .18 shows the two different
installation modes [I 1- 101.
Figure 11.I 8 Two dvyeleent modes of installationfor side airbags:

(a) seat-mounted; (b) door-mounted.
For lateral collisions, special attention must be given to the sensor installa-
tion. The sensors must be as close as possible to the doors; however, they also
must be located in the area where the passenger compartment remains intact,
even in severe accidents. The free distance between the side door sill and the
sensor should be as small and as stiff as possible to transmit the acceleration
data created by the impacting car to the sensor unit as rapidly as possible.
New sensors that measure pressure during contact with the impacting vehicle
are in production. (See also Section 10.2.2.4.)
1 1.2.2.2 Side Head Protection Airbags

After installation of the torso side airbag, solutions for head protection in
lateral collisions became more important. Figure 1 1.19 [ l 1 .11] shows an
analysis of severe injuries for different regions of the human body in lateral
collisions. The data were taken from cars before the head protection airbag was
installed. These data show that injuries are most common to the head, thorax,
and neck. Again, several installation modes for head airbags are possible, as
shown in Figure 1 1.20 [ll-41.
Meanwhile, the "inflatable curtain" has become an attractive installation

system. The head airbag is installed in the side roof area as a pre-assembled
unit to protect the heads of the outside front and outside rear passengers in
severe lateral collisions. Figure I 1.21 [ 1 1- 121 demonstrates deployment of a
typical window airbag, which prevents severe head impact against the vehicle
Head Neck Thorax Abdomen Pelvis Arm Leg
Figure 11.19 Severe injuries related to the human body in lateiaal

collisions. (Source: Ref: 11-11.)
Interrelationships
Figure 11.20 Different head airbag systems for lateral collisions.

(Source: ReL 11-4.)
Figure 11.21 Side window airbag. (Source: ReL 11-12.)

interior such as the B-pillar and the pop-out of the occupants' heads through
the side window area. As mentioned, the side head airbag also will be used
for the protection of passengers in rollover accidents.
1 1.3 Rear-End Collisions

For the development of vehicles related to their performance in rear-end
accident simulation tests, in addition to the full-scale test (check of the
integrity of the fuel tank and fuel lines), the resistance of the seats, and the
performance of the seat backs are tested most frequently in sled tests and in
tests that are defined by the vehicle manufacturer. The legal test procedures
for the seat and head restraints are valid for the single components but not
for total vehicle tests. Although the number of rear-end collisions is small
compared to the number of frontal and lateral collisions, the whiplash injuries
that occur in rear-end collisions are complicated and the injured people take
a longer period to recover. Rear-end collisions also represent a high percent-
age of cost with respect to insurance claims. Therefore, the newest features
for occupant protection in rear-end collisions also are based on the results of
accident statistics. In a study conducted by Temming [l l -131, it was found
that females have a higher risk compared to male occupants and that a large
number of injuries occur (see Figure 11.22) at a relatively low change in
velocity of approximately 8-12 kmlh (5-7.5 mph).
For evaluation of the performance of the seat and the seat head restraints,
a large number of mathematical models exist, and new types of three-
dimensional dummies have been developed. For example, many activities
conducted in Sweden [ 1 1 - 14 through 11- 161 resulted in the development
of the Bio-RID dummy. This dummy has a multi-segmented spine and
allows better determination of the performance of the seat, including the
head restraint. A paper by TNO Automotive [I 1-171 describes the RID 2-2
prototype, which was developed in the European whiplash project. The
results of tests with this dummy have improved the situation with respect to
the evaluation and assessment of the vehicle, the seat, and the head restraint
related to their performance in rear-end collisions.
The physical relationships among the partners in rear-end crashes are described
briefly here. In addition to some plastic deformation, the impacting car is
Interrelationships
Velocity change, Av (kmlh)
Figure 11.22 Relationship between the risk of whiplash injury and Av

(single rear-end impact, belted occupants). (Soulace: Re$ 11-13.)
creating a change in velocity of the impacted car, Av,, depending on the col-
lision speed and the vehicle weight.
Av, = v, . mi
m, + mi
where
Av, = Change of velocity of the crashed car

v,. = Relative collision speed
m, = Mass of the impacting car
m, = Mass of the crashed car
The acceleration level of the impacted vehicle usually is significantly below

the deceleration level observed in frontal collisions because the free deforma-
tion zone in the rear of the impacted vehicle is larger on average, and the relative
velocities are smaller compared to a car-to-car fiontal collision. Figure 9.13 shows
that even in a collision with a speed of48.3 kmlh (30 mph) and a heavy barrier
( 1 800 kg [3968 lb]), the maximum acceleration, measured at the impacted

car, is in the range of only 15g, compared with a frontal impact with a fixed
barrier of only 113 (see also Figure 9-4).
The forces created by the inertia force of the occupant are transmitted first
to the seat back, then to the seat frame, and (because of the rearward bend-
ing of the spine) to the headrest. Some basic requirements must be fulfilled.
It is evident that the seat back should not bend too much, so that the head
contact is so late that the head does not receive any support from the head
restraint. Another problem will occur if the headrest is too low. In this case,
it might increase the injuries to the occupants. Data are published from
some institutions (e.g., in Switzerland and also by the Insurance Institute for
Highway Safety (IIHS) [ l 1- 18]), which represent the results of head restraint
with respect to geometric data. Figure 11.23 demonstrates the measurement
methods and the methods of the judgment with respect to geometric data as
performed by the IIHS.
Backsel (cm)
Figure 11.23 Head restmint measurement. (Source: ReJ: 11- 18.1
As mentioned, in rear-end collisions we experience a dynamic loading of

the seat, the seat back, and the head restraint. Thus, the real performance is
shown only in dynamic tests. Because whiplash-associated disorders are so
significant in the field, car manufacturers have reacted on their own. One
indication of that activity is shown in the geometric layout of the head
restraints. Figure 11.24 demonstrates that the head restraints rated as "good"
Interrelationships
Head restraint
IM PROVEMENTS
Figzwe 11.24 Head restraint improvements.

(Source: Ref.' 11- 18.)
by IIHS increased from 3.3 to 46% between 1995 and 2003, and the head
restraints rated as "poor " decreased from 82 to 40%.
Another innovation was created by Saab [ l 1- 191. By active loading of the

seat back, the head restraint is brought into its most upward position. The
Saab active head restraint system (SAHR-System) has been in production
since 1997. Figure 11.25 shows its basic function.
Also, biomechanical activities such as the development of the Bio-RID dummy

have contributedto improvements of the seats, including the head restraints-not
only in laboratory tests but with respect to their performance in the field. If we
look to the neck extension angle as a function Av, there is significant improve-
ment in sled tests and on the roads. The analysis of accidents on Swedish roads
shows a reduction of 75% of accidents with people who received whiplash
injuries, after the SAHR-System was installed [ 1 1-20]. The study was performed
two years after the SAHR-System was in production.
Figure 11.25 Headrest developed by Saab.

(Sowce: R e t 11-19.}
1 1.4 Rollover Protection

The vehicle body with its designed structure, the vehicle interior, and the
glued-in front windshield provide good protection to vehicle occupants dur-
ing rollover accidents, particularly if the occupants are wearing seat belts.
For example, the glued-in windshield increases roof resistance during the
quasi-static roof test (FMVSS 2 16 [l l-2 11) by more than a factor of two
[ I 1-22]. Also, it is obvious that the doors of the vehicle should not open by
themselves during the rollover, and a valve activated by gravity forces in
the ventilation line from the fuel tank to the charcoal canister prevents fuel
leakage. Figure 1 1.26 shows head acceleration during a dynamic rollover
test, described in FMVSS 208, for two 50% male dummies on the driver and
passenger sides. As these data show, the resultant head acceleration is less
than 20 to 25g, which is far below the critical value of 8Og.
Interrelationships
- 30 -
C" 25
C
-
..*.
. . . .
. Passenger
Roof Roof Swell Swell Roof Roof Roof

driver passenger passenger driver driver passenger
side side side side side side
Figure 11.26 Dummy measzrrement data during a dynamic rolloveu:
For more severe accidents, the occupants' heads also must not come into
contact with other obstacles. For this reason, the airbag curtain, installed in
both sides of the vehicle roof, is a good supplement to the seat belt and the
side torso airbag. With a special sensor system that is designed to detect a
rollover situation, the airbag could be inflated rapidly enough to prevent head
injuries. Because the time for airbag inflation in a rollover situation could be
much longer compared to side impacts, it is not a big difficulty to use the side
window airbag (e.g., inflatable curtain) for rollover protection also [ll-231
(see Figure 1 1.2 1).
1 1.5 Special Requirements and Opportunities

for Sensor Applications
For the proper functioning of the pretensioners and airbags, inadvertent igni-
tion must be avoided, and the sensor system must show sufficient performance
during accidents. After several years of field experience, knowledge about
special effects has grown. In a Ph.D. study [l l-241, it was found that the use of
several sensors up front, plus central and other sensors used for side impacts,
could improve the determination of the severity of an accident. Even side
effects such as an increase of signals measured by the sensor accelerometer
due to sheet metal oscillations could be observed.
Another effect that has been taken into consideration is optimization of the
ignition timing of the airbag system. The following example shows the reason
for this statement. In a 40% offset crash against a deformable barrier, the
deceleration of the crashed car is not high enough in the beginning of
the crash to create a large forward movement of the occupant relative to the
vehicle. If the airbag inflates too early and the occupant hits the airbag at a
time when the gas in the airbag is already becoming cold and is leaving the
airbag through the pop-out valves and the inflator holes, then the occupant's
ride downward might be not adequate to prevent high deceleration of the
occupant [l l-251. For this reason, the layout of the sensor system must take
into consideration the total safety function of the car, including characteristics
of the body structure, the sensor system, the pretensioners, and the airbag
system. One conclusion of this analysis is that with a shorter ignition time,
the inflated airbag must keep the pressure for a longer period of time. This
consideration also is important for all types of pre-crash sensors because the
pressure in the airbag must be retained over a longer period of time.
Because some automated cruise control (ACC) sensors are used to control
the distance between the vehicle and the car in front of it, there is some hope
of using these same types of sensors, which are part of accident avoidance
systems, for devices in the field of mitigation of injuries. One big European
program for the analysis of various sensor systems is Chameleon [l l-261. In
this program, not only are different types of sensors used, as shown in
Table 11.1, but the possibilities of improving the protection systems in cars
for the fulfillment of legal requirements with respect to crash performance
are being investigated.
Interrelationships
TABLE 11.1
SENSOR SPECIFICATIONS PROVIDED BY
SENSOR SUPPLIERS [I1 -261
Saab IBEO TAMAM Temic Thales A.S.
Technology Microwave Laser-rotating Artificial Laser multi- Microwave
radar radar vision beam radar radar
Scan Rate 50 Hz 40 Hz 25 Hz 100 Hz 25 Hz
Delay Time 20 ms 25 ms 40 ms 10 ms 40 ms
Aperture 100" 270" 60" 3 x 15" 60"
Angle
Field Depth
Distance
Accuracy
Angle
Accuracy
Velocity *1 kph 10 kph +0.2
- kph
Accuracy
Figure 11.27 shows the possible function of the detection areas of various
types of sensors. With an analysis of the combined information, the failure
rate with respect to identification of the type of collision partners and the
severity of the accident could be reduced significantly [I 1-27].
One reason for this optimization process is the vision and hope to save lives
and to reduce severe injuries in the traffic environment. Some future possi-
bilities were demonstrated by DaimlerChrysler at the Frankfurt Motor Show
in 2001. In a visionary project for the future called PRE-SAFE, the company
mentioned several examples using pre-crash devices, such as the following:
Early pretensioning of the seat belt

Earlier ignition of the airbag propellant
Inclination of declined seat back and positioning of the headrest
The sunroof will be closed
The side door panel could be moved closer to the occupant
Although the technology for pre-crash sensor systems will be available rela-
tively soon, its introduction in production cars will depend on the benefit to
~;,.~,."Ihort
,.,., , , ,~
~ i ~ ~ ~ n ,!;!,!,!;!,!
c e distance
up to 14 m ::::-,:,:,- up to 1-5 m
Figure 11.27 Detection areas of variozis types o f sensors.

(Source: R e t 11-27.)
the customer and the reliability of the system. One basic requirement should
always be kept in mind: The positive effects of new systems should not by
themselves create even a single additional failure. Meanwhile, during 2002,
the DaimlerChrysler S-Class introduced pre-crash safety items, if a possible
accident is defined:
Pretension of the seatbelts

The seatback and cushion are brought into a normal position
The windows and the sunroof are closed
In the meantime, the PRE-SAFE system was extended by the DaimlerChrysler

PRE-SAFE braking system, where the brakes are applied automatically if an
accident is predefined. (See also Section 10.2.2.4 in Chapter 10.) Other car
manufacturers followed with similar functions.
Interrelationships
1.6 General Literature

Ziegahn, K.-F. "Airbag Symposium 2000," Fifth International Sympo-
sium and Exhibition on Sophisticated Car Occupant Safety Systems,
December 4-6,2000, Karlsruhe, Germany.
Ziegahn, K.-F. "Airbag Symposium 2004," Seventh lnternational Sym-

posium and Exhibition on Sophisticated Car Occupant Safety Systems,
November 29-December 1,2004, Fraunhofer Institut Chemische Tech-
nologie, Karlsruhe, Germany.
VDI Gesellschaft fiir Fahrzeug- und Verkehrstechnik: Fahrzeugkonzepte

f i r das 2. Jahrhundert Automobiltechnik, VDI-Berichte 1653, Dusseldorf,
200 1, ISBN 3- 18-091653-2.
VD1 Gesellschaft fur Fahrzeug- und Verkehrstechnik: Innovativer Kfz-

Insassen- und Partnerschutz, VDI-Berichte 1637, Dusseldorf, 200 1,
ISBN 3-18-091637-0.
VDI-Berichte 1794, Diisseldorf, 2003, ISBN 3-
VDI-Berichte 19 1 1, Dusseldorf, 2005, ISBN 3-
1.7 References
11- 1. Automotive Engineering, No. 4,2000, Society of Automotive Engi-
neers, Warrendale, PA, United States.
1 1-2. Author's information.
1 1-3. Audi AG. PROCON-TEN, technical description, 1989.
1 1-4. EnBlen, A., et al. "New Safety Elements in Volkswagen-Audi Auto-

mobiles," Lecture in Peking, 1996.
11-5. "Der Neue BMW 7er," ATZ/MTZ Extra, November 2001, Vieweg-
Verlag, Wiesbaden, Germany.
Insurance Institute for Highway Safety status report, Vol. 36, No. 10,
November 15, 200 1, Arlington, VA, United States.
Seiffert, U., et al. "Description of the Volkswagen Restraint Automatic

(VW-RA) Used in a Fleet Test Program," SAE Paper No. 740046,
States, I.-D., et al. "Volkswagen's Passive Seat BeltIKnee Bolster

Restraint (VW-RA): A Preliminary Field Performance Evaluation,"
Proceedings of the 21st Stapp Car Crash Conference, Society of
Automotive Engineers, Warrendale, PA, United States.
Holzner, M. "Virtuelle Prozessketten in der Funktionsauslegung des

Gesamtfahrzeuges," Proceedings of the ATZ/MTZ Congress, July
2002, Berlin, Germany.
Rabe, M., et al. "Seitenschutz mit Inflatable Curtain und Seiten-

Airbag," Conference held by Haus der Technik, March 23-24,2000,
Munich, Germany.
VW-GIDAS accident database, 2002, MAIS I+, belted struck side

passenger car occupant injuries in side collisions for different severi-
ties, multiple body regions possible. Unpublished. Volkswagen AG,
Wolfsburg, Germany.
AUTOLIV GmbH, Elmshorn, Germany, undated. Product informa-

tion material.
Temming, J., "Whiplash-Frequencies and Risk of Soft Tissue

Neck Injuries to Belted Occupants in Rear End Collisions of Pas-
senger Cars," Report of Volkswagen Corporate Research to BRITE-
EURAM-Project No. 3770 (Contract No. BRPR-CT96-022 I),
December 1997.
Davidsson, J., Svensson, M.Y., Flogard, A., Haland, Y., Jakobsson,

L., Lindner, A., Lijvsund, P., and Wiklund, K. "BioRID I-A New
Biofidelic Rear Impact Dummy," Proceedings of the International
Interrelationships
Research Council on the Biomechanics of Impact (IRCOBI) Confer-

ence, 1998 (a), pp. 377-390.
Davidsson, J., Flogard, A., Lovsund, P., and Svensson, M. "BioRID

P3-Design and Performance Compared to Hybrid I11 and Volunteers
in Rear Impacts at AV = 7 km / h, Proceedings of the 43rd Stapp Car
Crash Conference, 1999 (a), pp. 253-365, Society of Automotive
Davidson, J., Lovsund, P., Ono, K., Svensson, M.Y., and Inami, S.A.
"Comparison Between Volunteer, BioRID P3, and Hybrid I11 Perfor-
mance in Rear Impacts," Proceedings of the International Research
Council on the Biomechanics of lmpact (IRCOBI) Conference,
1999 (b), pp. 165-1 78.
Cappon, H., Philippens, M., and Wismans, J. "A New Test Method
for the Assessment of Neck Injuries in Rear End Collisions," Paper
No. 242, Experimental Safety Vehicle (ESV) Conference, Amsterdam,
The Netherlands, 200 1.
"Head Restraints, Status Report," Vol. 38, No. 9, Insurance lnstitute

for Highway Safety, September 25, 2003, Arlington, VA, United
States.
Saab GmbH. Product information material, undated. Saab Ope1

Sverige AB, Kista, Sweden.
Tschachtli, Chr. "Mechanismus gegen Peitschenhieb," Automobile-

Revue No. 25, June 2 I, 200 I, Bern, Switzerland.
National Highway Traffic Safety Administration (NHTSA) and

Federal Motor Vehicle Safety Standard (FMVSS), FMVSS 216, Roof
crush resistance, Washington, DC, United States.
Rabe, M. "The Role of Glazing for Car Safety," Proceedings of the

Glass Processing Days Conference, Tampere, June 6,2001, Society
of Automotive Engineers, Warrendale, PA, United States.
11-23. Meyer, I., et al. "Protection Technology 612001, Neues ijberschlag-

Schutzsystem," ATZ 108 (2001) No. 718, Wiesbaden, Germany.
11-24. Hubler, R. Unterstiitzung bei der Auslegung von Airbagsystemen

d u ~ FEM-Berechnungen,
h M-Verlag Mainz, 2001, ISBN 3-89653-
828-4.
1 1-25. Schwant, W. "Beurteilung des Schutzpotentials von Airbagsystemen

unter Beriicksichtigung unterschiedlicher Sensorzeiten," Conference
held by Haus der Technik Essen, "Kollisionsschutz im StraSen-
verkehr," February 29,2000.
11-26. Fiirstenberg, K.Ch., et al. "Development of a Pre-Crash Sensorial

System-the CHAMELEON Project," in VDI-Berichte 1653, Vehicle
Concepts for the 2nd Century ofAutomotive Technology, ed. by Ver-
ein Deutscher Ingenieure, VDI-Verlag, Diisseldorf, Germany, 2001,
pp. 289-3 lo.
11-27. Detlefsen, W., et al. "Evolution of Navigation to Cooperative Traf-

fic," in VDI-Berichte 1653, Vehicle Concepts for the 2nd Century
of the Automotive Technology, ed. by Verein Deutscher Ingenieure,
VDI-Verlag, Dusseldorf, Gemany, 200 1, pp. 357-372.
Pedestrian Protection
12.1 General
Although vehicle safety continues to improve, it is evident that if only pas-
senger car occupants become safer in the traffic environment, the total neces-
sary reduction of fatalities and severely injured people cannot be achieved.
This means that other groups, such as pedestrians, must be considered. If we
compare the situation in Europe in 1980, we could identify approximately
15,000 fatalities in pedestrian accidents, whereas in 1998 this number was
reduced below 7,000 fatalities [12- 11. Figure 12.1 shows that this trend is still
Figure 12.1 Fatal pedestrian injuries in Europe (EU). (Source: Re$ 12-2.)
positive, but the absolute numbers support the necessity to further improve
the protection of pedestrians [12-21.
The percentage of pedestrian fatalities with respect to all accidents in Germany

was 23.7% (3,095 out of 13,041) in 1980, and 12.2% (814 out of 6,618) in
2003. This means that not only the absolute number but the percentage of
fatal pedestrian accidents was reduced [12-21. From an analysis of accident
data in Germany, we can identify that the total number of injuries, seri-
ous injuries, and fatalities is highest in urban traffic, with a total of 37,581
(93.2%). This is followed by rural accidents with a total of 2,539 (6.3%),
and on motorways, with a total of 185 (0.5%). If we examine the number of
fatal accidents, that number is 983, with 67% in the city, 29% rural, and 4%
on motorways. Figure 12.2 shows the zone for the first contact in vehicle-
pedestrian collisions. Seventy percent of all first impact contact points are
in the front of the vehicle [12-21.
Figure 12.2 First contact points in vehiclepedestrian collisions [12-31.
The reduction of fatalities was achieved by vehicle improvements and other

measures in the traffic environment. Figure 12.3 supports that vehicle design
definitely has had a positive influence on safety. In Figure 12.3, we find the
frequency of pelvislfemur fractures for pedestrians older than 11 years old
[12-31 and a comparison between older cars (1974 to 1983) and newer cars
(from 1989). The improvements can be stated as being significant.
25
PEL VIS FEMUR
" Old c a n , accidents

from 1974 to 1983
(1 69 pedest~ians)
Recent c a n > 1989
(234 pedestrians)
I\dults 12-19 Adults 12-49

yea IS years
Figure 12.3 Frequency ofpelvis/femur~fi.acturesofpedestrians

older than 11years. (Source: Re$ 12-4.)
Figure 12.4 shows how complicated the accident situation can be in the real
world. This graph shows the head contacts on the front of the vehicle, as a
function of injury severity and collision severity. The accidents were analyzed
from 1985 to 2003.
There is a high distribution of contact areas on the vehicle in relation to the

Abbreviated Injury Scale (AIS). An in-depth study conducted by a university
team analyzed pedestrian accidents in the city of Hanover from 1985 until 2003,
with respect to injury frequency to various body regions [12-41. Figure 12.5
shows the results of the analysis.
This figure can be interpreted as follows: Head injuries occurred with 60.5%,
the windscreen area contributed to 19.0% of all injuries (16.6% of the head
injuries), the front hood 26.9% of all injuries (13.2% of the head injuries),
and the bonnet to 21.2% of all injuries (16.1% of the hip injuries). The
bumper contributed to 42.6% of all injuries (26.2% to the legs); the chest,
pelvic, and arm areas are involved by 44.7% of all injuries, and the leg area
by 77.5%. From these data, you also can see the high involvement of the
road (70.4%).
Figure 12.4 Head contact ofpedestrians on the,fi.ont hood o f the vehicle,

versus injury severity and collision speed. (Source: Re$ 12-4.)
Pedestrian hit by front of car

~ 6 5 5
frequency of injured
body regions
~ ~ i j u r ebody
d regio.1~rela:ed to car mpact area
. injuries caused by car
D ' . '
road 70.4%
Prof Dlpl hg 3 O!:r
Figure 12.5 Injury,jiequency to various body regions of pedestrian

and vehicle collisions. (Source: R e . 12-4.)
Also, contact with the road surface as a secondary impact causes different
severe injuries. Therefore, further improvements cannot be achieved by single
measures but only by the sum of several activities. Because of the mixed
traffic on the roads particularly in Europe and Asia-a large number of cars,
two-wheelers, and pedestrians-pedestrian protection has received great attention
from the public and, as a consequence, by legal authorities. Realizing that, in
many developing countries we have a dramatic change in traffic participants.
Thus, pedestrian protection might become even more important. Many actions
in the field of accident avoidance for pedestrians were mentioned in Chapter 6.
Future improvements for pedestrian protection must consider the possibilities
in the areas of both accident avoidance and mitigation of injuries. Figure 12.6
describes the task of the driver. Often, the critical situation of a potential
accident is underestimated by the driver. Likewise, pedestrians, especially
children, do not pay enough attention to other traffic participants.
-
Task:
Disturbances:
F. -1
' .Trafiic
I
I
.Weather cond~IM~s
-Road condions
-Light conditions .Other road users
Figure 12.6 The task of the driver: (Source: Rej.' 12-5.)
In Chapter 6 on accident avoidance with respect to the reduction of collisions

between pedestrians and other traffic participants, several measures were
recommended. The most important parameters are as follows:
View of the driver, also to avoid blind spots (a priority for truck drivers)
Detection of pedestrians, especially in dark environments
Traffic management and road design
Traffic signal improvement
Optimization of the red-light time on crosswalks for pedestrians
Measures on the vehicle, such as antilock brake systems (ABS) and daytime
running lights
The area (i.e., accident avoidance or mitigation of injuries) in which the level
of protection achieved is greater can be determined only in the future. We
also should keep in mind that the task is complex due to the different ages
and sizes of pedestrians. Directly related to age is the physical resistance of
the pedestrian; that is, older people have a higher risk of bone fractures.
The knowledge that pedestrian protection in Europe is of great importance

has stimulated intense research activities during recent years. The EEVC
with its Working Group 17 [12-61 finalized scientific work at the end of
1998. In this work, a good balance between test requirements and real injury
risk, new biomechanical knowledge, and the development of test experi-
ence, EEVC has created the basis for all current activities. For example, the
proposal introduced by the European Commission [12-71 and information to
the consumer provided by the European NCAP [12-81 are based on EEVC
activities. It has been some time since the NCAP findings presented the first
results in the area of pedestrian protection also. Although there is no direct
advantage for the vehicle owner and the self-protection ofthe car occupant, the
public interest related to pedestrian protection is increasing. In the European
NCAP-Test as of July 2006, we already find several cars with three stars and
with four stars related to pedestrian protection [12-91.
12.2 European NCAP-Test

The European NCAP-Test is performed on the basis of the activities of the
working group EEVC WG 17 [12-61 with some changes in the evaluation pro-
cedure. Figure 12.7 gives an overview of the Euro NCAP-Pedestrian Test.
Figure 12.7 Euro NCAP pedestrian test procedwe [12-81.
Table 12.1 describes the details of the test method and the evaluation criteria
for the original EEVC WG 17 and the Euro NCAP-Test.
12.3 Legislation Activities

After the intensive research phase by EEVC and in-depth studies done by vari-
ous laboratories and the industry, the European Commission and the Japanese
government decided to introduce pedestrian requirements. The introduction
date for the new type approval for phases 1 and 2 in Europe and the introduc-
tion in Japan are shown in Figure 12.8 [12- 101. The Euro NCAP-Test for
pedestrian protection has been published since 2001.
TABLE 12.1
EEVC WG 17 AND EURO NCAP PEDESTRIAN TEST DATA
EEVC Euro NCAP
Test Method WG 17 Lower Limit I Ueeer Limit
Lower legform impactor to 40 kmlh 40 kmlh
bumper 0" 0"
150 g
15"
6 mm
Upper legform impactor to 40 kmlh 40 kmlh
bumper 0" 0"
5 kN 5 kN 6 kN
300 Nm 300 Nm 380 Nm
Upper legform impactor to 2040 krnlh
bonnet leading edge 10-47"
5 kN 6 kN
300 Nm 380 Nm
Child headform impactor1
small adult headforrn
impactor to bonnet
Adult headform impactor, 40 krnlh
child headform to wind 65"
screen I65 mm
4.8kg
1000 1350
Adult headform impactor to 40 kmlh 40 krnlh
bonnet 65" 65"
I65 mm I65 mm
4.8kg 4.8kg
1000 1000 1350
Child headform impactor to 40 kmlh 40 kmlh
bonnet 50" 50"
130 mm 130 mm
2.5kg 2.5kg
1000 1000 1350
The European Commission introduced the EC-Directive 2OO3/lO2/EC related

to pedestrian protection in December 2003. The directive is relevant to all
vehicles of the class MI and derivatives N1 with a total weight of less than
2.5 tons. This directive requests tests with impactor-simulating human body
parts.
Figure 12.8 Introduction datesfor pedestrian protection

in Ezrrope and Japan.
For the type approval of vehicles as well as for the NCAP-Test, the funda-
mental work of EEVC WG 17 is the basis for these requirements. The tech-
nical parameters are described in the technical specifications 2004/90/EC,
February 2004. With head, leg, and hip impactors, different areas of the
front of the vehicle are tested. [n these tests, specified levels should not be
exceeded [12-111. Figure 12.9 [12-81 demonstrates the test procedure and
describes the child and adult headforms for phase 1.
The detailed requirements are as follows:

1.1 The following tests are required to be carried out; however, the
limit values specified in items 1.1.3 and 1.1.4 are required only for
monitoring purposes.
1.1.1 Legform to bumper: One of the legform tests described in

Section 1.1.1.1 or 1 . I . 1.2 is required to be performed:
Pedestrian Protection Test Procedures

according to EC-Directive 2003/102/EC
Adull headlonn lnnpnclor
Phase 1 4.8 kg 165 mm
ChM headfordmall
adult headform mpador
V p p r laglorm m m o r
L w e r leglorm Impad
Upper leglwm
-.
Figure 12.9 Pedestrian protection test procedure according

to 2OOYI 02/EC and 7O/lW E C .
1.1.1.1 Lower legform to bumper: The test is performed

at an impact speed of 40 kmlh. The maximum
dynamic knee bending angle shall not exceed 2 l.OO,
the maximum dynamic knee shearing displacement
shall not exceed 6.0 mm, and the acceleration
measured at the upper end of the tibia shall not
exceed 200g.
1.1.1.2 Upper legform to bumper: The test is performed at

an impact speed of 40 k d h . The instantaneous sum
of the impact forces with respect to time shall not
exceed 7.5 kN, and the bending moment on the test
impactor shall not exceed 5 10 Nm.
ChilcUsmall adult headform to bonnet top: The test is

performed at an impact speed of 35 k d h using a 3.5-kg test
impactor. The head performance criterion (HPC) shall not

exceed 1000 over two-thirds of the bonnet test area and 2000
for the remaining one-third of the bonnet test area.
Upper legform to bonnet leading edge: The test is performed

at an impact speed up to 40 kmlh. The instantaneous sum of
the impact forces with respect to time should not exceed a
possible target of 5.0 kN, and the bending moment on the test
impactor shall be recorded and compared with the possible
target of 300 Nm.
Adult headform to windscreen: The test is performed at an

impact speed of 35 k d h using a 4.8-kg test impactor. The
HPC shall be recorded and compared with the possible target
of 1000.
The European Commission and the various automobile companies also have
a voluntary commitment related to the improved pedestrian protection dis-
cussed. In the amendment to the Conch Directive 70/156/EC, the following
wording is written down:
On 21 December 2000, the Commission adopted a communi-

cation disctrssing the possibility of using a voluntary industry
commitment to improve the safety of pedestrians and other vzrl-
nerable road users as a ~eesultof collision with a motor vehicle.
In that communication the Conzmission agreed to undertake
discussions with the Ezr~opeanAutomobile Manufacturers Asso-
ciation (ACEA) with a view to reeach such a commitment and
to engage in parallel discussions with the Japan Automobile
Manz!factz~r~er*s
Association ( J A M ) and the Korea Automobile
Manzlfactz~rersAssociation (KAM).
In a comlnunication adopted on 11 Jzrly 2001 the Commission

presented the commitment undertaken by ACEA on pedestrian
protection to the Council and the European Parliament. The
commitment included a series of tests aimed at irnpreovingthe
constrzrction of thejiontal structures of'motor vehicles, as well
as a number oj'additional active and passive safety measures

also conducive to enhanced pedestrian protection. Those mea-
sures concerned: the equipment of all motor vehicles with
Antilock Brake Systems-ABS-as well as Daytime Running
Lights-DRL-, the ,fi~tzwe introduction of advanced active
safety ICTsystems and a voluntary ban on the sale ofrigid bull
bars. Since then, the J A M and KAMA have undeletaken com-
mitments similar to the one undertaken by ACEA, with the result
that 99% of the existing manzfacturers are covered by identical
commitments.
In the July 2001 communication, the Commission annozmced

that it would decide whether to accept the industry commitment
by means of a recommendation or to propose legislation in the
jield of pedestrian protection after having consulted the Ez~ro-
pean Parliarnent and the Council.
Both the European Parliament and the Council welcomed the

main elements of the industvy commitment as regards the mea-
suresfor improving the design of caqfronts. However; the Coun-
cil in its Conclusions of 26 November 2001 made its acceptance
of the commitment conditional to spec@ conditions including
the Member States involvement in the commitments' implemen-
tation and deferral of the adoption of DRL measures.
The European Parliament, in its resolution of 13 June 2002,

asked ,for a 'fframework" directive laying down the applica-
tion dates, the goals to be achieved and the method to monitor
their application. As a result, the Commission agreed to pro-
pose,frainework legislation, based on the relevant content of the
commitments.
This pi-oposal,for legislation was presented,for co-decision, and

the result was the Directive .XW/I O2/EC relating to theprotec-
tion of pedestrians and other vulnerable road users.
Phase 2 is still under discussion, but an initial set of requirements was published
by the European Commission, as can be seen from Figure 12.10 [12-81.
Figwe 12.10 Pedestrian protection phase 2, according

to 2OOYI 02/EC and 70/156/EC.
The following list shows the details of the phase 2 requirements:
1.2.1 Legform to bumper: One of the legform tests described in

1.2.1.1 or 1.2.1.2 is required to performed:
1.2.1.1 Lower legform to bumper: The test is performed

at an impact speed of 40 krnlh. The maximum
dynamic knee bending angle shall not exceed 19.0°,
the maximum dynamic knee shearing displacement
shall not exceed 6.0 mm, and the acceleration
measured at the upper end of the tibia shall not
exceed 170g.
In addition, the manufacturer may nominate bumper

test widths of up to 264 mm in total, where the
acceleration measured at the upper end of the tibia
shall not exceed 250g.
1.2.1.2 Upper legform to bonnet leading edge: The test is

performed at an impact speed up to 40 kmlh. The
instantaneous sum of the impact forces with respect
to time shall not exceed 7.5 kN, and the bending
moment on the test impactor shall not exceed
510 Nm.
1.2.2 Upper legform to bonnet leading edge: The test is performed

at an impact speed up to 40 kmlh. The instantaneous sum of
the impact forces with respect to time shall be compared with
a possible maximum of 5.0 kN, and the bending moment on
the test impactor shall be compared with a possible maximum
of 300 Nm.
This test shall be completed only for monitoring purposes,

and the results shall be fully recorded.
1.2.3 Childlsmall adult headform to bonnet top: The test is

performed at an impact speed of 35 kmlh using a 3.5-kg test
impactor. The HPC shall not exceed 1000 over two-thirds of
the bonnet area to be tested and 1700 for the remaining one-
third of this area.
1.3 The vehicle shall be equipped with a verified brake assist system.
Japan
A different set of requirements was chosen by the Japanese regulation of the
J-MLIT (Japanese Ministry of Land, Infrastructure and Transport) related
to pedestrian protection. The regulation is valid for passenger cars with
fewer than LO seats and its derivative light-duty trucks with a total weight
of 2.5 ton and the engine in front of the driver's seat. As can be seen from
Table 12.2, there also are component tests, but the tests differ, based on the
type of vehicle.
TABLE 12.2
PEDESTRIANTEST-REQUIREMENTS IN JAPAN [12-111
Child Head Adult Head
Definition Impactor (3.5 kg) Impactor (4.5 kg) Oliver Zander, BAST
N BLE Bonnet Impact Impact Angle Impact Impact Angle ["I
W
\D Height Angle Speed ["I Bonnet Speed Bonnet Top
Category Type [mm] ["I [kmlh] Top Area [kmlh] Area Limits
HIC 1000 1213 of whole bonnet test area]
1 Sedan <835 <30 32 65 32 65 HIC 2000 1113 of whole bonnet test area]
HIC 1000 1.213 of whole bonnet test area]
2 SUV 835 <30 32 60 32 90 HIC 2000 1113 of whole bonnet test area]
HIC 1000 1213 of whole bonnet test area]
3 1 BOX - >30
- 32 25 32 50 HIC 2000 1113 of whole bonnet test area]
12.4 Technical Solutions for Vehicles

in Accidents
The design trend for vehicles with respect to aerodynamic optimization, styl-
ing elements, and technical changes such as plastic front ends, including the
bumper, definitely helped to reduce the negative consequences in pedestrian
accidents. Also, the facts that the outside contour of the vehicle must have
only parts with a radius greater than 3.2 mm (0.13 in.) and that the outside
mirrors must fold away during an impact have reduced the number of injuries.
This also is true for the glued-in laminated windshield, which prevents head
injuries to a greater extent, especially cuts from glass separation.
The new European requirements are demanding many more changes for
vehicles. Some examples of possible solutions are described in the following.
It is evident that three areas in particular must be redesigned [12-101:
During the leg test, the front impact zone with its bumper system must
absorb the kinetic energy of the pedestrian leg impact device and must
prevent the rotation of the legform. Whereas the impact energies can be
reduced due to energy-absorbing components in the front bumper, the
rotation of the leg must be reduced by the geometric shape of the front of
the vehicle.
The hip impact test requires a redesign of the front hood, with the front
hood latch transverse bar.
The most complicated part of the requirements is the reduction of the

deceleration level during the headform impact to fulfill the head protec-
tion criteria (HPC) at the front hood of the vehicle. If we assume that
the impact speed is 40 kmlh (25 mph) and the average deceleration level
should not exceed 60g, then we already need a free deformation length
below the front hood of approximately I00 mm (4 in.).
In Figure 12.11, computer animation shows which vehicle components under

the front hood would have to be redesigned [12-I].
Because of the design restrictions and the location of vehicle components

under the front hood and to accommodate the requested free field of view
Figure 12.11 Engine compartment with an 80-mm (3.1 -in.)

lowered front hood. (Source: ReJ: 12- 1.)
for the driver, different design solutions are being investigated to ensure that
the front hood is not built too high. For example, if a pedestrian hits the
bumper, a mechanical or pyrotechnic upward movement of the front hood
could achieve the necessary free deformation area.
Another solution was demonstrated first at the International Automobile Show

in Frankfurt [12- 121. Figure 12.12 shows possible improvements in one spe-
cific vehicle, which were achieved by dynamic uplifting of the front hood rear
edge by approximately 100 mm (4 in.) and the use of pedestrian protection
airbags. Via two airbags on the left and the right sides at the rear part of the
front hood, it is lifted upward. The bags also improve the head impact situation
against the windshield. For all systems that must be activated by pyrotechnic
actuators or airbags, the sensor system must discriminate not only objects of
different sizes but objects of different stiffnesses as well. This means, for
example, that the sensor system must be able to differentiate an impact against
pedestrians or trees. It also is important that at low-speed impacts against
obstacles other than pedestrians, the system will not be activated. For this
reason, the priority lies in the development of reliable sensor systems.
Head-injury Death risk, Oh

criteria
30 W h ( I 9 mph) 40 k t r h (25 nph)
' Wthout the P A Wth the P A 1
Figure 12.12 Pedestrian protection via front hood airbags.

(Sozrrce: Ref: 12-12.)
12.5 References
Hahn, W. "Ful3gangerschutz-Von der Unfallanalyse zur Entwick-
lungsanforderung," Verband der Automobilindustrie (Association
of the [German] Automobile Industry) (VDA) Technical Congress,
200 1, Frankfurt am Main.
IRTAD International Traffic Safety Data and Analysis Group,

2006.
Hella AG, Technische Information FuBgangerschutz, Germany,

2006.
Otte, D. "Severity and Mechanism of Head Impacts in Car to Pedes-

trian Accidents," International Research Council on the Biomechanics
of Impact (IRCOBI) Conference, Sitges, 1999.
Hartlieb, M. "Ful3gangerschutz-Moglichkeiten und Potenziale

zukunftiger Technologien," Verband der Automobilindustrie (Asso-
ciation of the [German] Automobile Industry) (VDA) Technical
Congress, 200 1, Frankfurt am Main.
Kallikske, I. "Requirements for Pedestrian Protection at Motor

Vehicles and Solution Concepts," Proceedings of the Verein
Deutscher Ingenieure VDI-K Congress 2001. Also Friesen, F., eta].
"Optimierung von Fahrzeugen hinsichtlich des Beinaufpralltests,"
ATZ, Vol. 104, No. 5,2002, ISSN 0001 -2785, Vieweg-Verlag, Wies-
baden, Germany. Also European New Car Assessment Program
(EURONCAP), "New Testing and Assessment Protocol" (V 3.1,
November 2001). Also www.easi.de/training.
12-10. Pfaff, I., et a/. Frontendauslegung fiir den passiven Fuflgangerschutz,

Volkswagen AG, Wolfsburg, Germany, 2006.
12- 1 1. Zander, O., [email protected].
12.12 AUTOLIV GmbH. Press release, Internationale Automobil-

Ausstellung ( I M ) (International Automobile Exhibition), Frankfwrt
am Main, 200 1.
Compatibility
Because of progress in the field of mitigation of injuries for vehicle occupants,
more activities in areas other than the simulation of typical road accidents
(such as front, lateral, and rear impacts) have been investigated in greater
detail. One of these activities is the question of compatibility among traffic
participants and how that compatibility could be increased. Compatibility
in this context means the positive performance of traffic participants among
each other in the event of an accident. In the traffic environment, the follow-
ing group of participants must be considered: pedestrians, two-wheelers, and
occupants of passenger cars, trucks, and buses. Injuries could occur during
impact with another collision partner or during a single accident against a
fixed obstacle. For compatibility investigations, it has been shown that a
global approach, which includes all groups of possible collision partners, is
complicated, and the solutions are more than complex. Thus, in the follow-
ing analysis, this compatibility view is limited to passenger car, S W , and
light-duty truck collisions.
For the layout of the vehicle design, the following criteria have a direct influ-
ence on the collision process:
Mass ratio of the collision partners
Geometry of the vehicle structure
Force-deflection characteristic of the vehicle structures, stiffness
The location of the installation, size, and mass of the powertrain unit
The type and rigidity offered by the occupant cell (survival space)
The performance of the steering wheel and column, dashboard, knee

impact zone, and pedals during deformation
The following figures show some data relating to the relevant points
mentioned.
Vehicle mass. Figure 13.1 demonstrates for the U.S. market that the death
rate is different for the weight and the type of vehicles. In general, heavier
cars are safer, but cars generally are safer than pickup trucks. These data also
include the performance of the driver, showing not only the influence of the
vehicle design [l3- 1 1.
A - pickups
weight: 2.500 1b 3.500 4.500 5.500
Figure 13.1 Driver deaths per million registered vehicles by vehicle

weight, 1999 to 2003, models in the United States during 2000 to
2004 [13-I]. The rates are adjusted to accozmt,for some dijjferences
in driver age and sex within and between vehicle types. The
remaining differences in vehicle use patterns and driver
demographics may account for some of the death mte d@erences.
Compatibility
or less Curb weight

bsl
Figure 13.2 Mass distribution of cars in the United States and Germany.
The mass ratio in car-to-car collisions has declined during recent years, as
can be seen in Figure 13.2 [13-21.
From the diagram, one can determine that with a mass ratio smaller than 1.7
in Germany and below 2 in the United States, most car-to-car collisions are
covered.
Geometry of the vehicle structure. Because we also must consider lateral

car-to-car or light-duty-trucWSUV-to-car accidents, the geometry of the col-
liding cars has major importance to the behavior in these types of collisions
(see Figure 13.3).
Stiffness. Although the vehicle manufacturers are not publishing their data,
we find in Figure 13.10 some force-deflection curves for different car groups.
It is surprising that the differences in this test do not show a higher deviation,
at least in the first 600 mm.
Bottom U Pad581
Bumper Zone
(508 - 406 mm]
stralia Japan Europe US Australla Japan
Passenger Car SLN& LTV
Figure 13.3 Longitudinal member height [13-31.
Research activities on the subject of compatibility were performed as early

as 1970 [13-41. Because of other priorities, the intensity of research on
compatibility was relatively low but increased during the mid-1990s. Before
the theoretical analyses are discussed, we first should examine the relevant
accident data.
Figure 13.4 shows the distribution for car accidents with fatal and seriously
injured persons in Germany.
Figure 13.5 demonstrates the situation in the United States, showing the num-
ber of fatal accidents for car occupants and for SUV occupants [13-61.
Although some statistical differences in various countries are evident, one

could state that we still find a high number of single-car accidents and a rela-
tively high number in crashes where compatibility plays an important role
(e.g., car to car, car to pickup truck, car to truck, car to SUV). In the case of
SUVs, the percentage of single-car accidents is higher. Because of the much
lower market penetration of SUVs in Germany (new registrations in 2004
Compatibility
Fatal car occupants rn 2003 Seriously ~ n j u r e dc a r o c c u p a n t s In 2003
ISingle car Car to car

1Car to motorbike Ii Car to truck
Car to others B Car to more than 2 other road users
Figure 13.4 Car accident categories in Germany [13-51.
m$tribution of deathsin ciiis and mb 1

2000-03 models during 2001-04
Car occupant deaths SUV occupant deaths
15% crashes 10% crashes

tnvolving >2
involving +2
vehicleslother
blg lruck
or bus
8% car to suv \//

- Y'
16% car to car
Figure 13.5 Car and SUVjatal accidents in the United States [13-61.
approximately 5%, compared to 50% in the United States), car-to-SUV and

SUV-to-car accidents do not play such an important role compared to that in
the United States.
If we further analyze the type of accidents, the Volvo accident database [13-71
and GIDAS data [13-51 show the results displayed in Figure 13.6.
Crash Type Distribution Proportion of seriously or latally injured

Volvo Accident Database car occupants with respect to impact direction
'*?m
Other 5.796- Others
Dr~veOn Rwd - 6?i
Non-Struck Sidq
Frontal
Struck S~dey,
' 58%
24%
Sidem
-pI
Figure 13.6 Distribution of crash types [13-5, 13-71.
If we consider that the rear impact and the multiple collision often also include
car-to-car collisions, the frontal collision is still of high importance. However,
the lateral collision shows that we must consider both types of accidents when
examining compatibility.
Another important element is the relative velocity of vehicles with severe

injuries (MAIS 3+). Figure 13.7 shows that a v,,~,,,,, of approximately
170 kmlh (106 mph) again covers 90% of all accidents. With respect to
the overlapping ratio of the vehicle, the accident analysis data show that a
high percentage of accidents occurs with an overlapping of less than 50%.
Also, the deformation length of the vehicles in relation to severe and fatal
injuries shows a direct correlation. If cars are so heavily deformed that the
Compatibility
Closing Velocity [kmlh]
Figzrve 13.7 Accidentfiequency as a function o f closing velocity

deformation already reaches the windshield area, the probability of severe

injuries is high.
13.1 Theoretical Analysis

As mentioned, the legislation today mainly covers self protection, which
means the protection of the occupants in their own vehicles. If we compare
two vehicles where m, = 1000 kg (2205 Ib) and m2 = 1600 kg (3527 Ib) in
a vehicle against a fixed barrier collision and use triangular force deflection
characteristics, then we find a deformation length of 0.6 m (2 ft) in a 50-km/h
(3 1-mph) frontal barrier crash (plastic deformation). Figure 13.8 shows the
force deflection characteristics.
The maximum deformation force can be calculated by multiplying the mass

times the maximum deceleration, neglecting that a partial portion of the
vehicle mass is already at a standstill. The maximum deceleration is 33g, and
Deformation Force
[kNl
Deformation Length
Figure 13.8 Theoreticalforce deflection characteristics

in a crash against aJixed barrier
the maximum forces for the vehicle m, with 1000 kg (2205 lb) are 330 kN;
for the vehicle m2 with 1600 kg (3527 lb), the forces increase to 528 kN. Fig-
ure 13.9 illustrates the car-to-car collision, where two vehicles collide against
each other. The change of velocity can be calculated by the formulas
and
Compatibility
I Deformation force PN]
0.7 0.8 0.5 OA 0.3 02 0.1 0 0.1 02 0.3 0.4 0.6
Ddormation lenglh In]
Figure 13.9 Theoreticalforce deflection characteristic,

as a,firnction o f deformation length.
and
The deformation length of the lighter vehicle increases with this theoretical
=
observation to 0.74 m (2.43 ft). For the heavier vehicle, the deformation
length is reduced to approximately 0.46 m (1.5 ft). From this theoretical
example, we can easily follow the conclusion that this deformation must
be available in the light vehicle without disturbing the occupant cell. Both
the larger deformation and the higher g-level in the lighter vehicle must be
considered with respect of the layout of the restraint system.
The question of whether a compatible free-deformation characteristic could

be achieved for various vehicles is not answered easily. In a study performed
by Huibers [13-91, the force deflection characteristics for a number of cars
measured behind the deformable barrier element demonstrates interesting
results. Figure 13.10 shows the average force deflection characteristic for
various groups of cars. The four different vehicle groups (phases) of the
forceldeflection characteristic are shown. Number 3 shows subcompact
Figure 13.10 Force dejection characteristics for varioza vehicle groups.

(Source: ReJ: 13-9.)
cars sold in Europe, such as the Volkswagen Golf or the Toyota Corolla.
Number 4 shows larger vehicles, such as the Audi A6 or the Mercedes E 200.
Number 6 shows MPVs, such as the Volkswagen Sharan or the Chrysler
Voyager. Number 7 shows minis, such as the Volkswagen Polo or the
Peugeot 206.
It is interesting that for all passenger cars to a deformation length (offset test)
of 1 m (3.3 ft), the force deflection characteristic is very similar. This is not
the case for the MPVs. We believe this situation can be solved. This
requires a test or a computer simulation that eliminates the existing differences
of today. Keeping the preceding results in mind, we should concentrate on
the geometric layout of the vehicle structure and the strong integrity of the
vehicle compartment.
One approach was described by Schwarz and Zobel [13-41. They define a
bulkhead concept that requests a minimum resistance of the occupant cell,
together with a test for the agressiveness of the front end of a vehicle. The
Compatibility
front end could be checked, for example, in a crash against a special hydraulic
deformable barrier, where the energy absorption capability of the impacting
car versus this barrier could be used as a scale for compatibility.
This approach also helps in the design phase of a vehicle. Relou [13-101 has
developed a simplified mathematical model, where he simulates the front
structure by a mathematical model. This model is much easier to handle with
respect to computer power, cost, and time than the finite element model. The
different force deflection zones are reflected versus an optimal deformation
force distribution of the front structure of the vehicle. Figure 13.11 shows
the layout of the model.
Figure 13.11 Simulation model for frontal impacts.

(Sotlrece: Ref 13-10.)
In Figure 13.12, the actual data of the investigated vehicle are shown versus
the optimal compatibility index [KVG], which is a pre-defined force deflec-
tion characteristic. The left side of the figure shows the actual force versus
deformation, and the right side shows the compatibility index (KVG).
700
-Force A
600jl Force 01
Ref 11 \
'\
...... KVG 61
i s :
z
" 400
I
Displacement [mm] Displacement [mm]
Figure 13.12 Compatibility index versus deformation.

The advantage of this mathematical approach is the possibility of optimizing

by computer techniques the different vehicles of one car manufacturer or of
a group such as Ford, Volkswagen, or DaimlerChrysler, without performing
expensive car-to-car or car-to-barrier crash tests. 1t is evident that this method
also could be used to simulate lateral collisions. With these new tools, it is
possible to simulate a great portion of the real-world accident science and to
use this data for vehicle design.
13.2 Research Analysis Proposals, Consumer

Information, and Legal Requirements
In October 2005 in Dresden, Germany, nearly all worldwide actors discussed
the question of "compatibility on the road." In his conclusion, the conference
chairman showed the test and analysis procedures, as well as assessments
for compatibility. Figure 13.13 gives an overview related to the different
possibilities [13-111.
In the United States, we already find some legal activities that could be
described as follows [13-121:
Compatibility
Existing Proposals
Compatibility Test Procedures Compatibility Assessments
Full-W~dlhDeformable Barr~er VNT 1 HNT

Progresstve Deformable Barr~er Suffic~entSupport Test (SST)
Movmg Deformable Barr~er Impulse Assessments
Quasi-Stat~c Pressure Test Energy Assessments
Fln~teElement Analysis AHOF IHOF
Slal~cMeasurement 01 Slruclures PDB Aggresslv~tyMetrtc
Deformallon Llm~t
VNT = Vemcal component ol negawe lntrus~onAssessmenls
Icnamn horn largcl r u loaa
~
UNT = Honronlal crmwrenlol n q a h v e Dummy Assessmenls
dewatran Born larget row tom
PEAS = Pnmanl ancrgy absolbing structure Quasl-Stattc Mtnlmum Load
SEAS = Secondarv energy rDsomrg slruclum
AHOFIHOF =Average heghl d force PEAS ISEAS In Bumpe- 7---
PO8 = PrcqresstveOelmaDle bamer
Figzwe 13.13 Existing p~*oposalsvegal*dingtest and assessment p~*ocedures.
Industry self commitment:
Phase 1 : Ensures a common structural interaction zone and aims to avoid

override
Phase 2: Adds performance requirements for SEAS (secondary energy

absorbing structure)
Phase 3: Development of a dynamic assessment, proposals are FEM

(finite element method) simulations, full-width deformable
barrier test, and MDB (mobile deformable barrier) approach
Phase 4: Stiffness matching
The self commitment of the new light-duty trucks should comply at the begin-
ning of September 2009 with two options:
257
1. Low ride heights, primary energy-absorbing structure overlap with part

581 zone
2. Blocker beams, secondary energy-absorbing structure overlap with part

581 zone
Figure 13.14 shows the static requirements for energy-absorbing structures.
508 rnm1 1 406 rnm
Figure 13.14 Static ~~equivements

oj'enevgy-absorbing structuves (13- 121.
The NHTSA is looking via the AHOF (average height of force) approach to
optimize the stiffness of the vehicle with the following data:
Compatibility
AHOF over first 400 mm

- Low: Less than 408 mm
- Medium: 408 to 5 16 mm
- High: Greater than 5 16
Work stiffness for the first 400 mm

- lntegral of NCAP force-deflection curve, is really worldenergy
- Low: Less than 1300 Nimm
- Medium: 1300 to 1700 Nlmm
- High: Greater than 1700 Nlmm
Also in the United States, the IIHS barrier test (see Figure 13.15) is used in
frontal, lateral vehicle-to-vehicle tests, as well as for the NCAP-Test.
cai passenger car

llHS barrier taller and rides higher, contoured edges
-FM VSS 2 14 Barrier u llHS Barrier

Figure 13.15 Comparison of FMVSS 214 and the IIHS barrier [13-61.
As can easily be seen from the picture, the IIHS barrier is high above ground
and has rounded edges.
In Europe, two tests are analyzed. Figure 13.16 describes the TRL (Transport
Research Laboratory, U.K.) barrier test [13-21.
Figure 13.16 TRL barrier test.
Figure 13.17 shows the French approach.
Both evaluation procedures are not ready for the rulemaking procedure. Both
tests and the HOF (AHOF) assessment were analyzed by Zobel [13-111 with
the following results:
TRL Full-Width Barrier Test
There is a strong mass dependency due to the higher forces that gener-
ally occur with heavier vehicles. This results in, on average, larger errors
between measured and average values for larger cars.
The resolution of the LCW makes the evaluation sensitive to the impact
location of frontal structural parts.
Compatibility
Figure 13.17 Progressive deformable barrier-the French approach [13-21.
Smoothing cannot remove the influence of this sensitivity.
This sensitivity also may have implications for reproducibility, as all

vehicles have slight differences.
The deformation element detects only stepwise forces. This may mis-
represent the homogeneity of a vehicle in cases where the barrier is not
bottomed out.
PDB (Progressive Deformable Barrier) Assessment
Force homogeneity is not evaluated in the horizontal direction.
Horizontal members cannot be evaluated due to the turning of the vehicle

in an offset crash. This is in spite of the presence of a deformable element,
which aims to induce shear forces.
Stiffer cross beams are penalized.
Heavier vehicles are penalized.

All vehicles are encouraged to be softer.
All vehicles are encouraged to converge structures to the lower edge of

the barrier. It is obvious that small cars are more able to fulfill this than
SUVs.
HOF (Height of Force) Assessments
HOF and UWHOF are both unstable at lower force levels.
AHOF enforces an average lowering of forces, which increases the pos-

sibility for geometrical compatibility but does not enforce it.
AHOF does not assess force homogeneity.
13.3 Conclusions
Compatibility is a very strong element for the future of even safer cars on
the road. Based on the accident data, opinions differ about the benefit of
compatibility measures. The BAST [13-51 comes to the conclusion that 8%
of all fatally and 4.2% of all seriously injured car occupants in Germany will
benefit from compatible frontal car structures.
IIHS [13-61 predicted for a 2003 car front-to-front compatibility if SUV

front ends become more car-like, a 16% reduction for lighter SUVs (3000 to
3099 Ib), and a 29% reduction for heavier SUVs (4000 to 4099 lb).
Because of the already high number of crash tests used in the development
of the car, one of the tests already used should be modified to judge the car-
to-car performance as well.
An interesting alternative could be the use of FEM vehicles to check vehicle-

to-vehicle crash performance.
In the priority for the hture rule-making process, it is important not to hrther
increase self protection while reducing the protection of collision partners.
For example, because of a higher impact speed against a frontal barrier, if
the deformation force of the relevant vehicle must be increased, the vehicle
Compatibility
might become too aggressive. The other method, which is to make the crash
length of this relevant vehicle longer, is impossible because of space and
weight factors.
13.4 References
13-2. Biirkle, H. "Today's Relevance of Compatibility in Real World

Accidents," ATZ/MTZ, October 2005, Dresden, Germany.
13-3. O'Reilly, P. "Crash Compatibility-The Need for a Harmonized

Approach," ATZ/MTZ, October 2005, Dresden, Germany.
13-4. Appel, H. and Deter, T. "Crash Compatibility for Passenger Cars-

How to Achieve?" in VDI-Berichte 1471, Innovativer Kfz-Insassen-
und Partnerschutz, ed. by Verein Deutscher Ingenieure, VDI-Verlag,
Diisseldorf, Germany, 1999.
See also:
Appel, H. "Sind kleinere Wagen unsicherer als grofle? Unfallhau-
figkeit, Selbstschutz und Partnerschutz bestimmen die Sicherheit,"
VDI-Nachrichten, 1975, No. 7, pp. 14-1 6.
Bangemann, C. "Knall-Studie," Auto, Motol; Sport, 1999, No. 19,

pp. 42-49.
Brite Euram 4049: The Development of Criteria and Standards for

Vehicle Compatibility, Final Report 2001.
Digges, K., et al. "Stiffness and Geometric Incompatibility in Colli-

sions Between Cars and Light Trucks," SAE Paper No. 2003-01-0907,
Edwards, M., et al. "Compatibility-The Essential Requirements for

Cars in Frontal Impacts," International Conference on Vehicle Safety
2000, Institution of Mechanical Engineers, London, UK, June 7-9,
2000.
Farber, E. "EECV Research in the Field of Improvement of Crash

Compatibility Between Passenger Cars," 17th International Techni-
cal Conference on the Enhanced Safety of Vehicles, Amsterdam,
June 4-7,2001, U.S. Department of Transportation, Washington, DC,
United States.
Hackenberg, U., Rabe, M., and Friedewald, K. "Influence of Compat-

ibility on Car Design," VDI-Berichte 1471, Innovativer Kfz-Insassen-
und Partnerschutz, Tagung Berlin, September 30-October 1, 1999.
Proceedings of the 19th ESV Conference, June 2005, Washington,

DC, United States.
Schwarz, T. and Zobel, R. "Ermittlung der Zellsteifigkeit zur kom-

patiblen Auslegung von Pkw-Frontstrukturen," Crash-Tech 2000,
May 18-1 9,2000, Miinchen, TUV Akademie GmbH.
Seiffert, U., Hamilton, J., and Boersch, F. "Compatibility of Traf-

fic Participants," 3rd International Congress on Automotive Safety,
Vol. 1, July 15-1 7, 1974, San Francisco, CA, U.S. Department of
Transportation, United States.
Seiffert, U . "Hohere Aufprallgeschwindigkeit versus Kompatibil-

itat," Tagung Crash Tech Special, TUV ~ k a d e m i eGmbH, Miinchen,
March 9-10, 1998.
Zobel, R., el al. "Development of Criteria and Standards for

Vehicle Compatibility," 17th International Technical Confer-
ence on the Enhanced Safety of Vehicles, Amsterdam, June 4-7,
2001, U .S. Department of Transportation.
13-5. Pastor, C. "Potential Benefit of Improved Vehicle Compatibility,"

ATZ/MTZ, October 2005, Dresden, Germany.
13-6. O'Neill, B. "Crash Compatibility: An American View," ATZ/MTZ,

October 2005, Dresden, Germany.
Compatibility
Ivarsson, I. "Safety Strategy & Requirements," ATZ/MTZ, October

2005, Dresden, Germany.
Schwarz, T. "Selbst- und Partnerschutz bei frontalen Pkw-Pkw-

Kollisionen (Kompatibilitat)," Fortschritt-Berichte VDI Series 12,
No. 502, VDI-Verlag, Diisseldorf, Germany, 2002.
Huibers, J, and de Beer, E. "Current Front Stiffness of European

Vehicles with Regard to Compatibility," 17th International Techni-
cal Conference on the Enhanced Safety of Vehicles, Amsterdam,
June 4-7, 2001, U.S. Department of Transportation, National
Highway Traffic Safety Administration.
Relou, J. Methoden zur Entwicklung Crash-kompatibler Fahrzeuge,

Dissertation, Technical University of Braunschweig, Shaker-Verlag,
Aachen, Germany, 200 1, ISBN 3-8265-7804-X.
Zobel, R. "Stimulus Presentation," ATZ/MTZ, October 2005, Dresden,

Germany.
13-12. Scheef, J. Presentation, February 2004, Audi AG, 2004.

Computer Support for the
Development of Safety
Components
14.1 The Basics

The virtual development process has already become a reality in the research
and pre-development phases in all areas of the vehicle. This includes the area
of accident avoidance and the field of mitigation of injuries. This chapter
describes the tools for accident simulation. Together with the geometric data of
the vehicle components,an optimization process among the design, performance
during the accident, noise, and vibration must be performed. The breakthrough
in the field of calculation for safety items was achieved by the development of
stable software programs and the much higher performance of the supercom-
puter. One reason for the success was an interdisciplinary working group
of the German research organization FAT [ 14- 1 1. Figure 14.1, created by
Holzer et al. [14-21, shows how much progress in simulation techniques has
been achieved in the meantime.
Independent of the success of the simulation technique, it is very important

that the computation is supported by comparative analysis by using the results
of hardware tests. The performance of the supercomputer has doubled nearly
every two years, and many software programs running on computers can be
used by engineers without limitations. Although the number of parameters
that influence accident simulation tests has increased in recent years, calcu-
lation becomes an important part of the development process. In the newly
defined product creation process, the simulation and calculation technique
is an integrated part of marketing, development, production, finance, quality
assurance, and service.
t
Increase in
Demands
"
Sensor.
400
Famng Paris
Head Impact
Equlprnent and Integration
Rght-HandCIrive Crash o~~aurnm'as
-- -
Pob Cnsh -AZTIBwnpor T=t
.. 40% Ortsot Against D o l m b h Barrier
ECELaWral Crash * Rear.End Collirion Structure and Add.On-Pam
USLataral Crash
Onset Crash Against a Faxed Wall
FMVSS-ZOENS-NCAP b
Model Complexity
Figure 14.1 Simulation effort per car development verst~smodel

complexity. (Source: Re$ 14-2.)
14.2 Description of the Numeric Tools

The most important task for any computer-supported engineering work is the
generation and use of physical-mathematical models.
During the creation of the model, you must know the physical input data and
the vehicle component that shall be investigated. Depending on the status of
the project, the results achieved by calculation will be able to support trends
and to detail solutions up to full prognosis capability. A large number of
devices are already available to convert physical models into mathematical
ones.
For an automotive engineer, the development of new software is not as important

as its creative use. For a crash calculation, several calculation models are
used. These models include knowledge systems and a coupling of FEM and
MKS models together with the test database. This database must be updated
continuously. Today, material data and the connecting and the manufacturing
process of the material also are part of the calculation method.
The CAD (computer aided design) database is the basis for the use of FEM
(finite element model) calculation. Meanwhile, the performance of the total
Computer Support
vehicle and the performance of single components are optimized by com-

puter simulation techniques. Most automotive companies use both high-
power supercomputers and simplified models that can run on less-expensive
computers. In many applications, the performance of the vehicle structure
is calculated by using FEM crash software with multi-body systems, such
as MADYMO for the simulation of the occupant and restraint systems. The
solution used today is a software-based coupling of both systems. This solu-
tion allows the replacement of certain modules and an integrated calculation
of the total vehicle together with the restraint system (airbag and belts) and
with the choice of different sizes of occupants.
It is important that the scientific knowledge database is updated continuously

on the actual level for the optimization of certain vehicle components, as well
as for total vehicle performance.
14.3 Calculation of Components

In many accidents, the front longitudinal beams are important vehicle parts
for energy absorption. Numerous theories have been established by research
in the area of fold bulging as one means to achieve an optimal energy conver-
sion. In practical tests, the following formula for the average bulging force
for the profile shown was determined, as shown in Table 14.1.
where
-
OF = Dimensional coefficient
aF = Elastic limit of the material

a~ = 1.0 to 1.5 velocity dependent
S, = Sheet thickness
U, = Profile circumference with the sheet metal thickness S,
U, = Total length of the profile
TABLE 14.1
CROSS SECTIONS OF DIFFERENT LONGITUDINAL BEAMS
(SOURCE: REF. 1-1.)
Computer Support
Figure 14.2 shows the results of FEM calculations of a longitudinal beam

during the deformation phases.
Figure 14.2 Finite element model (FEM) o f longitz~dinalbeams.

(Source: Ref: 1-1.)
The program module DYNA 3D and PAM-Crash can be used effectively

in these types of calculation problems. For example, rectangular tubes of
different profiles (square, hexagonal, and octagonal) were investigated by
computer-supported calculation methods. The results demonstrate that hex-
agonal and octagonal tubes have a greater ability to absorb energy compared to
that of square tubes. In real-world crash situations, the design of longitudinal

beams and the transverse support in front of them are important in achiev-
ing the fold-bulging process of the beams. The fold-bulging process has the
greatest capability for energy absorption during deformation. In the event of
buckling of the longitudinal beams, the capability to convert the crash into
deformation energy is the lowest. Figure 14.3 provides an overview of the
calculation of deformation force versus crash time in a frontal collision for
the front end of a vehicle.
Force
I60
kN
120
80
40
Time
Figwe 14.3 Comparison of dejormation jorce as a,fi~nction of time

,for the,front end of a vehicle. (Source: Re$ 1-1.)
The calculation time for the crash simulation of a front end (not the total car)
by the use of a modern supercomputer such as a NEC SX6 could be reduced
to approximately 15 min.
The program PAM-STAMP [14.3] can even include the production process.
Figure 14.4 shows how many parameters can be used today for this type of
calculation. The program AUTOFORM simulates the material formation; the
effective plastic elongation is taken from the program PAM-STAMP. Together
with the material data, we can calculate the foreseeable design.
Computer Support
1 Allocation of the elements
I Calculation of the thickness of the weighted elements
1 Classification of the elements
Writing of the new material data
PAM-CRASH
0utput.pc
I optional
contro1.p~
I
Figure 14.4 Dutujow diagrum,for the simzrlation of nzechunical data.
(Source: Ref 14-3.)
Reference 14-4 shows the increased reliability in crash safety development

through numerical robustness analysis.
14.4 Total Vehicle Crash Computation

In the concept study, the design engineer could run first calculations with
specified key data, such as wheelbase, track width, powertrain location and
size, auxiliaries, and other vehicle components. This allows the design
engineer to perform a much better initial design study. Consequently, the
first prototypes have much better quality with respect to their performance in
crash tests. For the simulation tool, finite element model, of the total vehicle,
350,000 elements are used. This requires not only powerful computers (either
parallel computers or a supercomputer) but good tools for generation of the
finite element model. Figure 14.5 demonstrates the quality of the computer
simulation technique in showing a 56-km/h (35-mph) offset crash against a
deforrnable barrier.
Crash beginning
Figtire 14.5 Total car crash simulation.

Computer Support
The finite element model (FEM) could be used not only for the total vehicle
crash but for the detailed analysis also. In a Ph.D. thesis, Hiibler [14-51 simu-
lated the vehicle floor panel to determine whether vibrations could influence
the triggering of the airbag sensors. In this case, the FEM for the body floor
panel together with the longitudinal bars and the inner wheelhouse had to use
668,000 elements to demonstrate the expected effect.
Through the level achieved today, the use of calculations became an important
element in vehicle development. The necessary future activities for an even
greater use is the consideration of the vehicle body structure together with the
different types of occupants, restraint systems, and the interior of the vehicle.
With this, the dummy, or with a more human-like dummy, performance in
the vehicle environment could be better calculated.
14.5 Occupant and Restraint System

Simulation
With the simulation of the dummies by the use of MKS (multi-body systems)
and/or FEM, restraint system performance can be optimized. Using MKS as
a dummy substitution, the necessary calculation time is much less, compared
with the FEM dummy.
The MKS methods allow safety development primarily in three areas:
Principle and trend analysis

Plausibility of the system
Components and detail development
A typical application field is the optimization of systems. Figure 14.6 shows

one example. In this case, two different finite element models are used in a
European side impact test. The results for the acceleration of the upper and
lower spine are shown.
Meanwhile, MADYMO dummies of various sizes (through a child up to a

95% male) are used for accident simulation runs. This is possible even for
out-of-position situations. Figure 14.7 shows the result of a calculation of an
a YI ~IJ HI 1111 n n 41 m w, IMXI
Time [ms] Time [ms]
Figure 14.6 Finite element models (FEM),for side impact and acceleration
o f the upper and lower spine.
unrestrained dummy test in the passenger side for the head and neck, compared
to the measurement of two tests [14-61.
As we can see from the calculated and measured data, the match is sufficient.
In other cases, the trend of results could be reproduced but not the actual
numbers. Therefore, the development technology could be categorized into
the following groups:
Expanded MADYMO models with extensive use of finite element fea-

tures, including a coupling with the PAM-Crash model.
Computer Support
Comparison of the Simulation

Results with Test Data.
Mode Head on IP.
Load Parameter:
Head Acceleration, Resultant

Mode Head on IP.
Load Parameter:
Upper Neck Force, Shear (F,)

Mode Head on IP.
Load Parameter:
Upper Neck Moment, Flexion-
Extension (My)
Figure 14.7 Comparison of calculations with test data.

(Soutw: R e t 14-6.)
Coupled MADYMO and PAM-Crash, where MADYMO is simulating

the dummy and PAM-Crash is simulating the airbag and the vehicle.
Using full finite element models (FEM) for occupant, airbag, and vehicle.
Special attention must be given to the simulation of side impacts. In a study

of the optimization of vehicle stiffness distributions by means of numerical
simulation for the development of crash-compatible vehicle components by
Kersten [14-71, it was found that the PAM-Crash Euro-SID-1-FE dummy
gives a better representation of the dummy testing compared to the MADYMO
Euro-SID dummy.
14.6 Pedestrian Simulation Tests

Because of the importance of pedestrian protection and the on-going rule-
making process, pedestrian simulation tests are done for the whole pedes-
trian as well as for component testing. Figure 14.8 shows a simulation of a
pedestrian impact.
14.7 Summary
The simulation tools for evaluating the performance of the vehicle, its compo-
nents, its occupants, and other traffic participants with respect to automobile
safety have made a hndamental contribution to the product creation pro-
cess. The continuous further development of these tools also in the areas of
accident avoidance, software development, mechatronic systems, durability,
and comfort is more than necessary to further improve the safety of traffic
participants [14-91.
14.8 References
14-1. Raasch, I., Scharnhorst, T., and Schelkle, E. "Report of the FAT
Working Group Finite Elements," ed, by Verband der Automobilin-
dustrie (Association of the [German] Automobile Industry) (VDA),
Frankfurt a.M., Germany, undated.
14-2. Holzner, M., et al. "The Virtual Crash Lab: Objectives, Requirements,
and Recent Developments," in VDI-Berichte 141 1, Numerical Analy-
sis and Simulation in Vehicle Engineering, ed, by Verein Deutscher
Ingenieure, VDI-Verlag, Dusseldorf, Germany, 1998, pp. 27-52.
Computer Support
40 kmlh with deceleration or brake dive
40 kmlh with deceleration, but no brake

dive
40 kmlh with deceleration and brake

dive
Figure 14.8 Simzdation oj'a pedestrian impact against the front end o f a
passenger car traveling at 40 k m h (25 mph). (Sowce: Ref 14-8.)
14-3. Scholz, S.-P., et al. "Crash Computation in Consideration of the

Metal Forming Process," in VDI-Berichte 141 1 , Numerical Analy-
sis and Simulation in Vehicle Engineering, ed. by Verein Deutscher
Ingenieure, VDI-Verlag, Diisseldorf, Germany, 1998, pp. 195-2 14.
14-4. Giinther, F., et al. "Increased Reliability in Crash Safety Develop-

ment through Numerical Robustness Analysis," VDI-Bericht, 1846,
Dusseldorf, 2004.
Hubler, R. Unterstutzung bei der Auslegung von Airbagsyste-

men durch FEM-Berechnung, M-Verlag, Mainz, Germany, 2001,
ISBN 3-89653-828-4.
Siebertz, K., et a/. "Occupant Protection Assessment from Out-

of-Protection Models," in VDI-Berichte 14 11, Numerical Analysis
and Simulation in Vehicle Engineering, ed, by Verein Deutscher
Ingenieure, VDI-Verlag, Dusseldorf, Germany, 1998, pp. 11 1-1 32.
Kersten, R. "Optimization of Vehicle Stiffness Distributions," internal

scientific report, University of Eindhoven, The Netherlands, 2002.
Koch, W., ef al. "Comprehensive Approach to Increased Pedestrian

Safety in Pedestrian-to-CarAccidents," in VDI-Berichte 1637, Inno-
vative Occupant and Partner Crash Protection, ed. by Verein Deutscher
Ingenieure, VDI-Verlag, Dusseldorf, Germany, 200 1.
Verein Deutscher Ingenieure (ed.), Numerical Analysis and Simulation

in Vehicle Engineering (1998). VDI-Berichte 1441, VDI-Verlag, Dus-
seldorf, Germany, 1998, See also Verein Deutscher Ingenieure (ed.),
Numerical Analysis and Simulation in Vehicle Engineering (2000),
VDI-Berichte 1559,VDr-Verlag, Dusseldorf, Germany, 2000. See also
Verein Deutscher Ingenieure, "Numerical Analysis and Simulation
in Vehicle Engineering," VDI-Verlag, Dusseldorf, Germany, 2004,
ISBN 3-1 8-09 1846-2.
About the Authors
Professor Dr.-lng. Ulrich Seiffert currently is the acting chairman of WiTech

Engineering GmbH, speaker of the Center of Traffic Management, and a
lecturer at the Technical University of Braunschweig since 1982. Until the
end of 1995, he was a member of the board for research and development
of Volkswagen AG in Germany. He also holds a number of patents in the
safety area and was awarded several times for his work in the field of safety
engineering. He is the author of a large number of books and has published
more than 300 technical papers.
Dr.-Ing. Lothar Wech is the general manager of TUV SUD Automotive

GmbH, TUV Siiddeutschland Group. He is competent in the field of passive
safety of vehicles and has served as chairman of the Crash-Tech Conferences
since 1992.
INDEX
Index Terms Links
Abbreviated Injury Scale (AIS) 73 74t 227

Acceleration 77 82 140f 146f
276f
and braking 43
vehicle dynamics in 6 56
Accident avoidance (active safety) 5 25
acceleration and braking in 43
adaptive cruise control and 47
environment and 5 6 26
information systems and 6 63
brake-by-wire in 52
brakes in 48
comfort and ergonomics in 29
definition of 3 4f
EEC rules for 13t
German rules for 13t
human factors in 5 25
U.S. rules for 14t
United Nations ECE rules for 13t
vehicle dynamics in 6 56
Accident rates, and age 22f 23f 24f
Accidents
car-to-car, simulation tests 133
criteria influencing 245
This page has been reformatted by Knovel to provide easier navigation.

Index Terms Links
Accidents (Cont.)
distribution of types 250f
fatalities in 18f 19f 20f 21f
24f 138f 225f 246f
249f
multiple-vehicle, fatalities 21f 24f
pedestrian
fatalities of 225f
first contact points in 226f
simulation test devices 94
technical solutions for 240
restraint system sensors and 172
rollovers
fatalities in 21f
simulation tests 145
simulation tools 267
single-vehicle, fatalities in 19 21f 22 24f
steering column impact 91 93f
total vehicle crash computation 273
vehicle body data 111f
see also Frontal impacts; Lateral
impacts; Rear impacts; Rollovers
Active chassis 33f
Active chassis control 60
Active devices, definition of 5
Active safety. See Accident avoidance
Actuators, definition of 5
Adaptive belt force limiter 154 156f 194 195f
Adaptive cruise control (ACC) 40 47
Advanced frontlight system (AFS) 44f

Index Terms Links
Aerodynamic coefficient 56 58f

Age, and accident rates 22f 23f 24f
Air conditioning 30
Airbags 160 198 201f
ignition timing of 218
knee protection 171 172f 200
pedestrian protection 171 242f
rollover protection 171
safety effectiveness of 163f
side impact systems 202 206
typical deployment strategy 173f
Alcohol 27
All-around view 37 38 40
All-wheel drive 45f
Aluminum space frame 114
Antilock brake system (ABS) 43 45f 51
Audi
A2 and A8 114
AG 192 193f
SUV-Q7 49
Australia
IHRA activity 16
NCAP tests 8t
number of safety rules 12
Austria, fatality rates 18f 20f 225f
AUTOFORM 272
Automatic stability control plus traction
(ASC+T) 44
Average height of force (AHOF) 258 262
Axial force 84

Index Terms Links
Axles 56 57f
rear axle lift coefficient 56 58f
Barényi, Bela 109

Barriers
fixed 134
IIHS 259f 260
progressive deformable 261
test 130
TRL 260
Belt clamps 153 154f 189 191
192
Belt forces 157 191f
Belt tensioners 153 154 155f
Bending moment 84
Bending torque, maximum 80t
Beru RDKS 60
Bicycle accident rates 18f 21f 26f
Biofidelity dummy 102
Biomechanical limits on humans 75f
Biomechanics 71
Bio-RID dummy 100 102f 212
Blood alcohol content (BAC) 27
BMW 700 series 33
Board frame 124
Brake assistant 48 50f
Brake-by-wire 52

Index Terms Links
Brakes 48
antilock 43 45f 51
basic requirements for 49
electrical wedge 56
electronic mechanical brake system
(EMB) 55
sensotronic brake control (SBC) 53
Bulging force 269
Bulkhead concept 254
Canada 8 16
CCS sensors 168f
Chalmers University 100
Chameleon 218 219t
Chassis, active 33f
Chassis control 59t 60
Chest injuries 80
Chest protection 83
performance criteria for 86
Child dummies 99 101f
Child restraints 179
Child seats
regulations for 181t
sensor identification of 175f 176
Climate control 32
Closing velocity 251f
Clutch, Haldex 46f
Collisions. See Accidents; Frontal
impacts; Lateral impacts; Rear impacts
Index Terms Links
Comfort, in accident avoidance 29

Communication systems 63
Compatibility of traffic participants 245
existing proposals regarding tests 257f
theoretical analysis of 251
Computer support for component
development 267
Computer-aided design (CAD) 268
Conch Directive 70/156/EC 234f 235 237f
Concussion 76
Convertibles 115 121 148 171f
Crashes. See Accidents; Frontal
impacts; Lateral impacts; Rear impacts
Criteria
chassis control 59t
head protection 77 78f 82
in rule-making process 82
viscous 80 82f 83
Crush resistance 124
Customers 2f 9
DaimlerChrysler
A-Class 112 113f 114
PRE-SAFE brake 47 49 177 179
S-Class 220
Data flow diagram 273f
Deceleration 71 74 76f 78f
128f 132f 189 190f
191f
Index Terms Links
Definitions 3
Deflection, lateral rip 83
Deformable crash barrier 130 131f
Deformation 189 190f
Deformation force 128f 129 272f
Denton 100
Doors 200
Drag coefficient 56 58f
Driver assistance systems 25 40 43
development of 61f 63f
Drivers 11 25
comfort of 29
fatalities of 246f
tasks of 229f
Dummies
biofidelity 102
Bio-RID 100 102f 212
child 99 101f
Euro-SID 97 143
Euro-SID-FE 278
finite element method 103 105f
frontal impact 97
Hybrid III 95f 96f 98
lateral impact 97
modeling of 102
multi-body 103 104f 105f
rear impact 97
Bio-RID 100 102f
rollover 97
three-dimensional 94

Index Terms Links
Dummies (Cont.)
two-dimensional 34
U.S. side impact (US-SID) 97 143
see also Test devices
DYNA 3D 271
Dynamic vehicle simulation tests 127
Economic Commission for Europe

(ECE) 12 16 141
EC-Directive 2003/102/EC 232 234f 237f
EC-Directive 2004/90/EC 233
ECE-R44 179 180f
rules for accident avoidance 13t
EEVC 230
WG 17 231 232t 233
EG 96/27 141f
Elastic rebound 127
Electrical wedge brake 56
Electro-hydraulic brake system (EHB) 53
Electronic control unit (ECU), airbag 161 162
Electronic differential system (EDS) 44 46
Electronic mechanical brake system (EMB) 55
Electronic Stabilisation Programme (ESP) 56 58 60
Electronic suspension system 33f
Elk avoidance maneuver 56
Environment, in accident avoidance 5 6 26
Ergonomics, in accident avoidance 29
ERTICO 64
ESP-Plus 58
Index Terms Links
European Automobile Manufacturers

Association (ACEA) 235
European Commission 232 235 236 237f
European Economic Community
(EEC) 12
NCAP-Test 8t 230 231f 232t
233
pedestrian fatalities 225f
European side impact dummies (Euro-SID) 97 143 278
European Space Agency (ESA) 64
Experimental Safety Vehicles (ESV) 71
Extension 79f 277f
External injuries 74
External safety, definition of 3
Facial injuries 74
FAT 267
Fatalities 18f 19f 20f 21f
24f 138f 225f 246f
249f
Feet 89
Field of view 37
Finite element method (FEM) 103 105f 268 269
273
longitudinal beam model 271f
side impact model 276f
test simulation 152f
Flexion 79f 277f
Index Terms Links
FMVSS 105 49
FMVSS 201 91 141 191f
FMVSS 208 82 160 163 198
216
test requirements of 165f 167f 168t
FMVSS 214 139 140f 141 145
259f
Force deflection 81f 252f 253f 254f
Four-wheel drive 44 45f
France, fatalities in 18f 20f
pedestrian 225f
Frontal impacts 97
airbags for 160
deformation force in 272f
design requirements for 136
force deflection in 81f
interrelationship of elements in 187
with pedestrians 279f
repair cost test for 117f
simulation tests for 127
simulation model for 255f
see also Accidents; Lateral impacts
Rear impacts; Rollovers
Fuel tanks 112 145 147f
g-Levels 71 74 75f 76
86 87 89
Gender, and accident rates 22 24f

Index Terms Links
Germany
accident rates 22 23f 25 26t
249f
customer interest in safety 9
FAT 267
fatalities in 17 18f 19 20f
249f
pedestrian 225f 226
GIDAS data 250
GPS 43
Head impact test devices 91 92f

Head injuries 74 76 210f
Head injury criteria (HIC) 77 78f 82
Head motion 197f 277f
Head–neck torque 79f
Head performance criterion (HPC) 235 238 240
Head protection, side airbags for 210
Head protection criteria (HPC) 77 78f 82
Head restraints 214f 215f
Head-up displays 36
Headlamps 41 44f
Heat loading 30 31f
Height of force (HOF) 262
Highway Loss Data Institute (HLDI) 8
Holloman Air Force Base 72
Hoods, pedestrian contact with 228f 240 242f
Index Terms Links
Human body part test devices 91

Human factors in accident avoidance 25
Humans
biomechanical limits on 75f
modeling of 102
see also Occupants
Hybrid gas generator 199f
Hybrid III dummy 95f 96f 98
Industry self commitment 256

Inflatable curtain 210
Information Society Initiative in
Standardization (ISIS) 64
Information Society Technologies
(IST) 64
Information systems 1
and accident avoidance 6 63
Injuries
Abbreviated Injury Scale (AIS) 73 74t 227
by accident type 249f
chest 80
external 74
facial 74
head 74 76 210f
internal 76
leg 210f
mitigation of 5 6 218
definition of 3 4f
neck 84 210f
Index Terms Links
Injuries (Cont.)
pelvic 210f
spinal 79
thoracic 210f
tolerance limits 73
whiplash 212 213f 215
Instrument panel 36f
Insurance Institute for Highway Safety
(IIHS) 8t 214 215 262
barrier test 259f 260
side impact tests 143 144f
Integrated safety system (ISS) 62
Intelligent Car Initiative 64
Intelligent roads 65
Interior safety, definition of 3
Internal injuries 76
International Automobile Show 241
International Experimental Safety
Vehicle Conference 7
International Harmonized Research
Activities (IHRA) 16
International Research Council on the
Biomechanics of Impact (IRCOBI) 71
ISOFIX seat 181 182f
Italy
fatalities in 20f
pedestrian 225f
IHRA activity in 16
IVHS 64

Index Terms Links
Japan Automobile Manufacturers

Association (JAMA) 235
Japan
fatalities in 18f 20f
IHRA activity in 16
NCAP tests 8t
pedestrian protection in 233f 238 239t
Knee protection
airbags for 171 172f 200
Korean Automobile Manufacturers
Association 235
KVG 255 256f
Laser scanner 39f

Lateral impacts
airbag systems for 169 202 206
dummies for 97
FEM model 276f
force deflection in 81f
interrelationship of elements in 200 202
requirement phase-in 144f
tests

Index Terms Links
Lateral impacts (Cont.)

IIHS 143 144f
requirements for 138
simulations 138
theoretical analysis of 202
see also Accidents; Frontal impacts;
Rear impacts; Rollovers
Lateral rip deflection 83
Leg protection, performance criteria for 87
Leg injuries 210f
Legislation
blood alcohol content 29
product liability 9
safety 11 231 251
early 7
Lift coefficient, rear axle 56 58f
Lighting 36 41 43
Load limitations 72 75f 80t
Load limiter, seat-belt 194
Loading conditions 84
Longitudinal beams 111 114 136 137
269 272
cross sections of 270t
FEM model of 271t
Longitudinal member height 248f
Low-speed impact, requirements for
testing 117

Index Terms Links
MADYMO 103 269 275

Man-machine interface 33 62
Mass, vehicle 246
Mathematical models 268
Mechanical belt tensioner 153 155f
Mechanical belt pretensioner 192
Mercedes SL 115
Microwave sensor concept 39f
Mirrors, outside 38 39f
Mitigation of injuries (passive safety) 5 6 218
definition of 3 4f
MKS 275
Modeling, human and dummy 102
Models
frontal impact 255f
longitudinal beam 271f
physical 268
see also Simulations
Moose avoidance maneuver 56
Motor Vehicle Safety Act 11
Motorized two-wheelers
accident rates 26f
fatalities 18f 21f
Multi-body dummies 103 104f 105f
Multiple-vehicle crashes, fatalities in 21f 24f

Index Terms Links
National Highway Traffic Safety

Administration (NHTSA) 141 157 160 258
Navigation systems 63
NCAP-Test 8t 164 230 231
232t 233
Neck injuries 84 210f
Neck shear force 86f 277f
Neck tension load 85f
Netherlands
fatalities in 20f
pedestrian 225f
governmental safety goal 7
New Car Assessment Program
(NCAP) 8t
Noise 30 31f 32
Numeric tools 268
Obstacle identification systems 38

Occipital condyle 80t 84
Occupants
protection of 151
simulation of 89 91 275
unrestrained 157 187 275 276f
velocity of 203f
see also Humans
Outside mirrors 38 39f

Index Terms Links
Overall Abbreviated Injury Scale

(OAIS) 73
PAM-Crash 271 276

PAM-STAMP 272
Passenger cars, fatalities in 18f 21f
Passengers, and accident rates 22f 26f
Passive devices, definition of 5
Passive restraint systems 196
Passive safety 5 6 218
definition of 3 4f
Patrick, L. 74 76
Patrick curve 76f
Pedestrians
accident rates of 23f 26f
accident simulation test devices 94
body contact with vehicle 227 228f
European NCAP-Test 230 231f 232t 233
fatalities of 18f 21f
head contact with hood 228f
pelvis/femur fractures of 226f
protection of 89 90f 171 225
phase-in of requirements 233f
test procedure for 233 237
simulation of 278 279f
Pelvic injuries 210f
of pedestrians 226f
Pelvic protection, performance criteria for 87
Index Terms Links
Performance criteria, for rule-making

process 86
Photonic Mixing Device 175
Physical models 268
Pole test 132 139 140f 141
Portugal, pedestrian fatalities 225f
Post-crash safety 6
Pre-crash 6
definition of 3
sensors and 219
PRE-SAFE brake 47 49 177 179
220
Pressure sensor 49
Pretensioners 189 191 192
Primary collision, definition of 5
PROCON-TEN 192 193f
Product liability legislation 9
Progressive deformable barriers 261
Propulsion slip control (ASR) 44 45f
Protection 6
chest 83
child, side airbag 208f
head
criteria for 77 78f 82
side airbags for 210
knee 87 171 172f 200
leg, performance criteria for 87
of occupants 151

Index Terms Links
Protection (Cont.)
of pedestrians 89 90f 171 225
phase-in of requirements 233f
test procedure for 233 237
pelvic, performance criteria for 87
rollover 141
side, airbags for 169 171f
thorax 206
Pyrotechnic belt tensioner 154 155f
Pyrotechnic pretensioner 192 194f
Quasi-static tests, requirements for 119
Radar 47
Rain sensor 38f
RAMSIS 36f 103 105f
Reaction force 129 130f
Reaction time 48
Rear axle lift coefficient 56 58f
Rear impacts
dummies for 97
Bio-RID 100 102f
RID 2-2 212
repair cost test 118f

Index Terms Links
Rear impacts (Cont.)

simulation tests of 145
Lateral impacts; Rollovers
Relative movement 188
Repair cost tests 117f 118f
Restraint systems 151
child 175f 176 179
comparison of 200 201f
definition of 5
passive 157 159f 196
sensors for 172
simulation of 275
see also Airbags; Seat belts
RID 2-2 212
Roads
conditions of 26t
fatality rate by type 18f
intelligent 65
Roadsters 115
Rollovers
dummies for 97
dynamic 217f
fatalities in 21f
protection in 216
side curtain airbag and 171
simulation tests of 145
Lateral impacts; Rear impacts
Roof, strength of 121 122f

Index Terms Links
Rotational acceleration 77
Rpm sensor 53
Rules for accident avoidance 13
Run-through buckle 157f
Ryan, James J. 72
Saab active head restraint system

(SAHR-System) 215 216f
SAE International 64 71
SAE J211/1 rev. March 95 84
Safety components
computer support for development of 267
calculation of 269
Seat anchorage point test 119
Seat belts 152
anchorage point test 119
clamps 153 154f 189 191
192
load limiter 194
passive 157 159f
pretensioners 189 191 192
retractors 154 193 194
tensioners 153 154 155f
three-point 152 160 189 197f
201f
two-point 158
usage data 196f
Seat track 121f
Seating reference point 35
Index Terms Links
Seats 189 212 207f

Secondary collision, definition of 5
Self-carrying body design 111 113f
Sensors 40
airbag 168f 169f 198
applications of 178f
definition of 5
detection areas of 219 220f
in exterior mirrors 38 39f
firing decision matrix 177f
microwave concept 39f
opportunities for 218
pre-crash 174 219
pressure 49
pyrotechnic pretensioner 193
radar 40 47
for restraint systems 172
rpm 53
sidewall torsion 60 61f
specifications for 219t
speed 45f
steering wheel angle 53
tire pressure 60
in vehicle compartment 174
vehicle surround 40f
yaw rate 53
Sensotronic brake control (SBC) 53
Severity index (SI) 76

Index Terms Links
Shear force 84
neck 86f 277f
Side impacts. See Lateral impacts
Sidewall torsion sensor 60 61f
SID-H3 98
Simulations
and car development 268f
of ergonomic design 35
frontal impact 255f
occupant 89 91 275
of pedestrians 278 279f
total car crash 274f
see also Models
Single-vehicle crashes, fatalities in 19 21f 22 24f
Smart restraints, definition of 5
Software 268
Spain, pedestrian fatalities 225f
Speed, accident rates 48f
Speed limits 19
Spinal injuries 79
Sport utility vehicles 46
Stapp, John 71
Stapp Car Crash Conference 71 72
Static forces, maximum 80t
Static requirements for energy-absorbing structures 258f
Steering column impact 91 93f
Steering wheel 137
angle sensor for 53
average distances to chest 188f
Stiffness 109 110f 259

Index Terms Links
Submarining 158 197

Suspension system, electronic 33f
SUVs 248 250 262
fatalities by accident type 249f
Sweden 7 12
Switzerland, fatalities in 18f
Test devices
head impact 91 92f
pedestrian impact 94
three-dimensional 34
torso impact 91 93
see also Dummies
Tests
dynamic vehicle simulation 127
frontal collision simulation 127
lateral impact 138
NCAP 164 230 231 232t
233
pole 132 139 140f 141
quasi-static, requirements for 119
repair cost 117f 118f
seat anchorage point 119
seat-belt anchorage point 119
of vehicle performance 56 59t
Thoracic Trauma Index (TTI) 86
Thorax
protection of, side airbags for 206
injuries of 210f
Index Terms Links
Three-dimensional dummies 94
Three-dimensional test device 34
Three-point seat belts 152 160 189 197f
201f
Tibia index (TI) 87
Tire pressure, requirements for 60
Torque
bending, maximum 80t
head-neck 79f
Torso impact test devices 91 93f
Total vehicle crash computation 273
Traffic guidance systems, development of 61f
Traffic management 64
Transatlantic Economy Dialog 16
Transport Research Laboratory (TRL) 260
TRL barrier test 260
Trucks, fatalities 21f
Two-dimensional dummies 34
Two-point seat belts 158
U.S. side impact dummy (US-SID) 97 143

UFS sensors 168f
United Kingdom
fatalities in 18f 20f
pedestrian 225f
IHRA activity 16

Index Terms Links
United Nations Economic Commission

for Europe (ECE) 12 13t 16 141
179 180f
United States
customer interest in safety 9
Department of Transportation 64
fatalities in 18f 19 20f 246f
249f
IHRA activity in 16
industry self commitment 256
Motor Vehicle Safety Act 11
NCAP tests 8t
seat-belt usage data 196f
Unrestrained occupants 157 187 275 276f
V-close 172
Vehicle body 109
and low-speed impact 117
energy absorption of parts 137f
external forces on 116f
geometry of 247
legal requirements for 116f
roof strength requirements 121 122f
seat anchorage requirements 119
seat-belt anchorage requirements 119
side structural reinforcements 143f
side structure requirements 122
Index Terms Links
Vehicle body (Cont.)

stiffness of 200 247
tests without occupants 119
velocity of 203f
Vehicle compartment
sensors in 174
Vehicle dynamics, in acceleration and
braking 6 56
Vehicle mass 246
Vehicle performance 26
minimum 43
tests of 56 59t
Vehicle safety
driving forces for 7 196 229
external, definition of 3
governmental goals for 7
interior, definition of 3
legal requirements for 115 116f
legislation for 7 11 231 251
number of rules for 12
passive 3 4f 5 6
218
post-crash 6
subdivisions of 3 4f
see also Accident avoidance
Velocity 128f 132f 133 139
140f 146f 190f 203f
change in 213f 252
closing 251f

Index Terms Links
Virtual development process 267

Viscous criteria (VC) 80 82f 83
Volkswagen 46
Beetle 111 112f
EOS 148 149f
Golf 158
Passat 111 113f
passive restraint system 197
Volvo 206
accident database 250
C70 170 171f
Weather conditions 30 31f

Wheel load 32f
Wheel-road contact 49 51f
Whiplash injuries 212 213f 215
Windscreen 136
Wiper systems 37 38
Working-place driver 33
World-SID 100f
Yaw moment controller 58

Yaw rate sensor 53

Automotive Safety Handbook PDF

Uploaded by

Copyright:

Available Formats

Automotive Safety Handbook PDF

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

Automotive Safety Handbook PDF

Uploaded by

Copyright:

Available Formats

Automotive Safety

For permission and licensing requests, contact:

Library of Congress Cataloging-in-Publication Data

Copyright © 2003, 2007 SAE International

This book on automotive safety describes the relevant development of safety

 'H¿QLWLRQV ......................................................................................... 3

3. Driving Forces for Increased Vehicle Safety ................................. 7

4. Safety Legislation ............................................................................11

5. Accident Data ................................................................................. 17

6. Accident Avoidance ........................................................................ 25

7. Biomechanics and Occupant Simulation ..................................... 71

7.5 Criteria in the Rule-Making Process ......................................... 82

8. Vehicle Body ................................................................................. 109

9. Dynamic Vehicle Simulation Tests .............................................. 127

9.1.3 Design Requirements of Frontal Collisions................... 136

10. Occupant Protection .................................................................... 151

11. Interrelationships Among Occupant,

11.4 Rollover Protection.................................................................. 216

12. Pedestrian Protection .................................................................. 225

13. Compatibility ................................................................................ 245

14. Computer Support for the Development of Safety

Index ...................................................................................................... 281

About the Authors ................................................................................ 293

For this reason, it is not surprising that in marketing questionnaires related

Figure I . 1 Conflicting demands by vehicle buyers.

Vehicle safety can be subdivided into three main areas:

Some additional definitions related to vehicle safety are as follows:

Accident avoidance. (Colloquial = Active safety.) All measures that

Mitigation of injuries. (Colloquial = Passive safety.) All measures that

Pre-crash. All measures to reduce the severity of an accident after the

External safety. Design of the external parts of the vehicle to reduce

Interior safety. Design of the vehicle interior parts to prevent additional

Traffic routing Protection of Protection Protection of

Restraint systems. Vehicle components (e.g., seat belts, airbags) that

Smart restraints, sensors, and actuators. These apply to occupant

Primary collision. Collision of the vehicle with another obstacle.

Secondary collision. Collision of the occupant with vehicle parts.

The two main areas of safety-accident avoidance and mitigation of inju-

In the field of accident avoidance, three main influences are found:

1. The human being

1. Protection during the crash

2-2. Braess, H.-H. and Seiffert, U. Handbuch Kraftfahrzeugtechnik, Vieweg-

Legislation certainly has a strong influence on safety. In Germany, with the

The vision of government activities related to safety is very strong (basis

The Netherlands and Sweden: Zero fatality vision

Canada and the European Union: 30% reduction in fatalities by 2010

A second effect was the activity initiated by consumer advocates such as

the results on various websites. Because of consumer interest for information

Although not discussed in public intensively, product liability legislation also

Highway Loss Data Institute, www.carsafety.org.

Klanner, W. "Status Report and Future Development of the Euro NCAP

3-4. Crashworthiness rating reports, various issues, sporadically published

3-5. "Safety Steps in the Spotlight," Automotive News, March 6, 2000.

The different government legislative institutions that are responsible for

1. Traffic routing (e.g., road design and traffic signals)

With respect to automotive legislation on a worldwide basis, there are differ-

Figure 4.1 Number of rules for vehicles in various countries.

In Europe, the situation with respect to the rule-making process is as follows.

and their light sources

'H¿QLWLRQV ......................................................................................... 3