The future of intelligent transport systems

249 Pages·2020·9.339 MB·English

by  George J. Dimitrakopoulos, Lorna Uden, Iraklis Varlamis

The Future of Intelligent
Transport Systems
George Dimitrakopoulos
Lorna Uden
Iraklis Varlamis
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2020 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, elec-
tronic or mechanical, including photocopying, recording, or any information storage and retrieval
system, without permission in writing from the publisher. Details on how to seek permission,
further information about the Publisher’s permissions policies and our arrangements with organi-
zations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found
at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experi-
ence broaden our understanding, changes in research methods, professional practices, or medical
treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluat-
ing and using any information, methods, compounds, or experiments described herein. In using
such information or methods they should be mindful of their own safety and the safety of others,
including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of products
liability, negligence or otherwise, or from any use or operation of any methods, products, instruc-
tions, or ideas contained in the material herein.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
ISBN: 978-0-12-818281-9
For information on all Elsevier publications
visit our website at https://www.elsevier.com/books-and-journals
Publisher: Joe Hayton
Acquisitions Editor: Brian Romer
Editorial Project Manager:
Aleksandra Packowska
Production Project Manager:
Punithavathy Govindaradjane
Designer: Victoria Pearson
Typeset by Thomson Digital
Introduction
The term Intelligent Transport Systems (ITS) was coined several years ago,
reflecting the continuously modernized manner in which people, vehicles, and
other objects of the transportation infrastructure move and communicate. Espe-
cially with the enormous advances and incorporation of Information and Com-
munication Technologies (ICT), ITS has become the cornerstone of transport
and attracts immense research interest from the academia, resulting in innova-
tive technological developments from the industry.
This book falls exactly in the realm of the above-mentioned grounds, con-
taining a holistic approach on the latest technological advances that transform
transport systems of all kinds and shape the way people travel around.
The contents of the book revolve around a set of pillars, namely:
1. technology enablers;
2. users;
3. business models;
4. regulation, policies, and standards; and
5. the future of ITS.
Overall, the book covers, in a holistic manner, aspects that are relevant to the
next generation of ITSs in one place. The presentation of the book will follow
the figure above in five parts.
Part 1 introduces the reader to all the technological enablers for building
ITS. It provides a holistic approach to intelligent transportation. In particular,
it will cover (a) the sensing technologies that can be used for data collection,
(b) the wireless communication advances that enable fast data transfer, and (c)
the computational technologies such as cloud and edge computing that allow
flexibility and push applications, data and computing services to the logical ex-
tremes of a network. Last, this part will also provide information on connected
vehicles and the available relevant test beds.
Part 2 of the book covers all aspects that are relevant to the users, as part
of the transportation chain. It explores the needs, preferences, and identifies
changes in travel decisions and technology acceptance.
Part 3 of the book focuses on the business and revenue models that influence
ITS. In particular, it discusses on the design and pricing of ITS related services,
the financing and revenue allocation models, the legal requirements and the
user (driver/passenger) rights, defines the value chain, investigated the financ-
ing schemes for new ITS concepts, and investigates revenue allocation models.
Part 4 will investigate all policies that affect ITS, as well as proposing new
methods to model ITS processes, so as to end up with new appropriate policies
for promoting technological advances, rather than hindering them.
The last part of the book focuses on ITS applications, which are present in
two different perspectives: (1) from the point of the transportation network and
xiii
xiv Introduction
the applications that can improve network safety, traffic flow, and also on smart
cities and urban mobility concepts, (2) from the vehicle point, with empha-
sis on autonomous driving. Emergency vehicle notification systems, variable
speed limits, dynamic traffic light sequence, collision avoidance systems are
some of the applications that will be presented in detail. Concerning autono-
mous driving, the various levels of autonomy, starting from the “eyes off” level
and moving to the “driver off” case will be presented. All existing technology
enablers for each level will be covered and the maturity of each solution will be
described. User acceptance and ethics issues will be presented in detail in order
to assist researchers, students, and practitioners to better design their solutions
in the future in order to achieve wider acceptance.
As such, the book is a unique resource where the reader can turn to study
EVERYTHING about ITS that is related to the future of mobility, combining
personalized mobility, big data, and autonomous driving.

Chapter 1
Sensing and perception
systems for ITS
1.1 Introduction: highly automated vehicles and the
importance of perception
The ever-increasing utilization of vehicles along with the ongoing immense
research in novel vehicular concepts has brought about the concept of highly
automated and autonomous vehicles. The automation of vehicles—ultimately
aiming at fully autonomous driving—has been identified as one major enabler
to master the Grand Societal Challenges “Individual Mobility” and “Energy Ef-
ficiency”. Highly automated driving functions (ADF) are one major step to be
taken. One of the major challenges to successfully realizing highly automated
driving is the step from SAE Level 2 (partial automation) to SAE Levels 3
(conditional automation), and above. At Level 3, the driver remains available
as a fallback option in the event of a failure in the automation chain, or if the
ADF reaches its operational boundaries. At higher levels (4 and 5), the driver
cannot be relied upon to intervene in a timely and appropriate manner, and con-
sequently, the automation must be capable of handling safety-critical situations
on its own. This is shown in Table 1.1.
The automation of vehicles is strongly linked to their interconnection (V2V
communications), as well as to their connection to the transportation (and also
telecommunication) infrastructure (V2I), as those kinds of communications can
pave the way for the design and delivery of innovative services and applications
supporting the driver and the passengers (cooperative, connected automated
mobility—CCAM). Despite the numerous advances in several initiatives related
to CCAM, there are still plenty of limitations to be overcome, especially in the
following areas:
1. Deployment cost reduction: At this time, CCAM solutions are associated
with high costs that are associated with the distribution of the necessary
infrastructure for their deployment.
2. Communication availability improvement for CCAM: Availability of state-
of-the-art communication infrastructure/technologies nation-wide.
3. Vehicle cooperation improvement: In-vehicle intelligence, connectivity, and
coordination among heterogeneous technologies.
The Future of Intelligent Transport Systems. http://dx.doi.org/10.1016/B978-0-12-818281-9.00001-2
Copyright © 2020 Elsevier Inc. All rights reserved. 3
4 PART | I ITS technology enablers
TABLE 1.1 Summary of levels of driving automation for on-road vehicles.
SAE Steering and System
level Name acceleration Perception Fallback capabilities
Human in charge of perception
0 No Driver Driver Driver None
automation
1 Driver Driver + Driver Driver Some driving
assistance System modes
2 Partial System Driver Driver Some driving
automation modes
System full in charge of perception
3 Conditional System System Driver Some driving
Automation modes
4 High System System System Some driving
Automation modes
5 Full System System System All driving
Automation modes
Source: ERTRAC, 2015.
4. Driving safety improvement: CCAM solutions that will assist the driver in
effectively handling sudden or unforeseen situations, especially for SAE
Levels 3 and beyond.
5. Business models: Solutions that will envisage new revenue generators for all
involved stakeholders, that is, vehicle-to-business communications.
6. Traveler’s information enhancement: Real-time, accurate, and tailored in-
formation provision to the driver, especially when information originates
from multiple sources and is associated with large amounts of data.
Last, while many prototypes exist, which demonstrate CCAM technologies,
they are confined to special applications and somehow limited to simple scenari-
os. Past and on-going projects on CCAM focus on vehicle platooning, where ve-
hicles operate in a well-defined and structured environment (highway scenarios).
In such a context, the vehicle needs to efficiently (in a fail-operational man-
ner) perceive its environment, that is acquire contextual information, so as to be
fully aware of its surroundings and be able to take optimal decisions regarding
its velocity, direction, and overall behavior on the road.
Any mobile robot must be able to localize itself, perceive its environment,
make decisions in response to those perceptions, and control actuators to move
about (Burgard et al., 1999). In many ways, autonomous cars are no differ-
ent. Thus many ideas from mobile robotics generally are directly applicable to
highly automated (also autonomous) driving. Examples include GPS/IMU fu-
sion with Kalman filters (Thrun, Burgard, & Fox, 2005), map-based localization
(Dellaert et al., 1999), and path planning based on trajectory scoring (Kelly &
Sensing and perception systems for ITS Chapter | 1 5
Stentz, 1998). Actuator control for high-speed driving is different than for typi-
cal mobile robots and is very challenging. However, excellent solutions exist
(Talvala, Kritayakirana, & Gerdes, 2011).
However, the general perception is unsolved for mobile robots and is the
focus of major efforts within the research community. Perception is much more
tractable within the context of autonomous driving. This is due to a number of
factors. For example, the number of object classes is smaller, the classes are
more distinct, rules offer a strong prior on what objects may be where at any
point in time, and expensive, high-quality laser sensing is appropriate. Never-
theless, perception is still very challenging due to the extremely low acceptable
error rate.
1.2 Driver’s sensor configurations and sensor fusion
Driven by the demand for fewer accidents and increased road safety, the auto-
motive industry has started with the implementation of driving assistance sys-
tems into vehicles several years ago. These assistance systems include adaptive
cruise control, blind-spot detection, forward collision warning, and automatic
emergency braking, among others. As main sensors for monitoring of the ve-
hicle environment 2D cameras were used, in recent times also RADAR sensors
have been increasingly employed for increased reliability. During the last years,
it became more and more evident that the imperfectness of capturing the vehicle
environment was one major limitation, often leading to system fail to func-
tion or to system switch off through auto detection. Particularly critical weather
situations (snow, ice, rain, fog) and certain object properties (e.g., small-sized,
nonreflecting, or transparent or mirroring obstacles) can lead to unreliable be-
havior. Also, mutual interference with other vehicles’ active sensor units cannot
be neglected with increasing penetration of deployed assistance systems.
Driving assistance is the first level of autonomous driving. Recent research
efforts address higher levels of driving autonomy (Fig. 1.1 and Table 1.2), going
beyond pure driver assistance systems toward fully autonomous driving, that
FIGURE 1.1 Evolution in ADF capabilities across SAE levels.
6 PART | I ITS technology enablers
TABLE 1.2 Levels of automated driving defined by VDA J3016 and key
performance figures for autonomous driving (Level 3+ requires advanced
fail-operational dependability and ASIL D safety level).
Automa- Functional Driver Perception Safety
tion level description interaction redundancy Dependability level
Level 0 No High None Fail-silent QM
automation
Level 1 Driver Medium– Complemen- Fail-silent ASIL
assistance High tary A or B
Level 2 Partial Medium Combining Fail-safe ASIL B
automation
Level 3 Conditional Moderate Partially Fail-safe ASIL C
automation overlap
Level 4 High Seldom Largely Fail-operational ASIL C
automation overlap (single error) or D
Level 5 Full None Fully overlap Fail-operational ASIL D
automation (single error)
is, VDA/SAE Level 3+. This involves fail-operational behavior and the highest
levels of safety (ASIL D).
It is a common understanding that reliability improvement and advanced
solutions for environmental perception (prerequisites for autonomous driving)
can only be achieved by sensor diversity combined with data fusion approaches,
due to the physical limitations of single sensor principles.
In the automotive domain (according to all major OEMs), robust and reli-
able automated driving will only be achievable by combining and fusing data of
three different sensor systems: LiDAR, Radar, camera, exploiting their specific
strengths as depicted in Tables 1.3 and 1.4.
McKinsey predicts an overall share of 78% for processors (37%), optical
(28%), and RADAR sensors (13%) in 2025 (Table 1.5) among automotive semi-
conductors, reflecting the main electronic components of highly automated ve-
hicles as announced by OEMs.
This is evidenced not only in market reports but also in the technology road-
maps of major OEMs. Strategy Analytics has analyzed the sensor demand for
environmental acquisition and indicates high annual growth rates for RADAR,
LIDAR, and 2D camera sensors for the coming years.
However, currently available solutions for highly automated driving have
not reached readiness levels suitable for the automotive industry. Although sys-
tem deployment costs for these demonstration vehicles are very high, this is
acceptable and normal for novel low-TRL technologies. However, the inability
to achieve fail-operational levels is a significant roadblock to their adoption.