Professional Documents
Culture Documents
RITA 2018 Proceedings of The 6th International Conference On Robot Intelligence Technology and Applications Anwar P. P. Abdul Majeed
RITA 2018 Proceedings of The 6th International Conference On Robot Intelligence Technology and Applications Anwar P. P. Abdul Majeed
RITA 2018 Proceedings of The 6th International Conference On Robot Intelligence Technology and Applications Anwar P. P. Abdul Majeed
https://textbookfull.com/product/proceedings-of-international-
conference-on-computational-intelligence-and-data-engineering-
proceedings-of-iccide-2018-nabendu-chaki/
Advances in Computational Intelligence Proceedings of
Second International Conference on Computational
Intelligence 2018 Sudip Kumar Sahana
https://textbookfull.com/product/advances-in-computational-
intelligence-proceedings-of-second-international-conference-on-
computational-intelligence-2018-sudip-kumar-sahana/
https://textbookfull.com/product/proceedings-of-international-
conference-on-artificial-intelligence-and-applications-
icaia-2020-poonam-bansal/
https://textbookfull.com/product/proceedings-of-6th-
international-conference-in-software-engineering-for-defence-
applications-seda-2018-paolo-ciancarini/
https://textbookfull.com/product/data-analytics-and-learning-
proceedings-of-dal-2018-p-nagabhushan/
https://textbookfull.com/product/proceedings-of-the-
international-conference-on-information-technology-systems-
icits-2018-1st-edition-alvaro-rocha/
Lecture Notes in Mechanical Engineering
RITA 2018
Proceedings of the 6th International
Conference on Robot Intelligence
Technology and Applications
Lecture Notes in Mechanical Engineering
Lecture Notes in Mechanical Engineering (LNME) publishes the latest
developments in Mechanical Engineering - quickly, informally and with high
quality. Original research reported in proceedings and post-proceedings represents
the core of LNME. Volumes published in LNME embrace all aspects, subfields and
new challenges of mechanical engineering. Topics in the series include:
• Engineering Design
• Machinery and Machine Elements
• Mechanical Structures and Stress Analysis
• Automotive Engineering
• Engine Technology
• Aerospace Technology and Astronautics
• Nanotechnology and Microengineering
• Control, Robotics, Mechatronics
• MEMS
• Theoretical and Applied Mechanics
• Dynamical Systems, Control
• Fluid Mechanics
• Engineering Thermodynamics, Heat and Mass Transfer
• Manufacturing
• Precision Engineering, Instrumentation, Measurement
• Materials Engineering
• Tribology and Surface Technology
To submit a proposal or request further information, please contact the Springer
Editor in your country:
China: Li Shen at li.shen@springer.com
India: Dr. Akash Chakraborty at akash.chakraborty@springernature.com
Rest of Asia, Australia, New Zealand: Swati Meherishi at
swati.meherishi@springer.com
All other countries: Dr. Leontina Di Cecco at Leontina.dicecco@springer.com
To submit a proposal for a monograph, please check our Springer Tracts in
Mechanical Engineering at http://www.springer.com/series/11693 or contact
Leontina.dicecco@springer.com
Indexed by SCOPUS. The books of the series are submitted for indexing to
Web of Science.
Junmo Kim
Editors
RITA 2018
Proceedings of the 6th International
Conference on Robot Intelligence Technology
and Applications
123
Editors
Anwar P. P. Abdul Majeed Jessnor Arif Mat-Jizat
Faculty of Manufacturing Engineering Faculty of Manufacturing Engineering
Universiti Malaysia Pahang Universiti Malaysia Pahang
Pekan, Pahang Darul Makmur, Malaysia Pekan, Pahang Darul Makmur, Malaysia
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Organising Committee
General Chairs
Programme Chairs
Organising Chairs
v
vi Organising Committee
Workshop/Tutorial Chair
Publications Chairs
Exhibition Chairs
Secretariat
ix
x Preface
The editors hope that readers find this volume informative. We thank
Springer LNME for undertaking the publication of this volume. We also would like
to thank the conference organising committee for their hard work in realising the
conference.
xi
xii Contents
Dr. Mohd Hasnun Arif Hassan earned his first degree in mechanical engineering
from the Technische Hochschule Bingen in Germany in 2010. He then pursued a
master’s degree in mechanical engineering at the University of Malaya in Kuala
Lumpur, graduating with distinction in 2012. After that, he embarked on his Ph.D.
at the Universiti Malaysia Pahang (UMP), where he studied the head injuries
sustained by soccer players due to heading. He completed his Ph.D. study in 2016
and then continued to serve UMP as Senior Lecturer. He is currently Director of the
Innovative Manufacturing, Mechatronics and Sports Laboratory (iMAMS), which
was founded by Prof. Zahari Taha, who was his supervisor for his doctoral study.
His research interests include finite element modelling of the interaction between
human and sports equipment, instrumentation of sports equipment and injury
xv
xvi About the Editors
prevention particularly with regard to sport. His work aims to apply engineering
principles in sports to not only enhance the performance of an athlete, but also
prevent injuries.
Prof. Zahari Taha graduated with a B.Sc. in aeronautical engineering with hon-
ours from the University of Bath, UK. He obtained his Ph.D. in dynamics and
control of robots from the University of Wales Institute of Science and Technology
in 1987. He is currently Professor of Manufacturing Engineering at Universiti
Malaysia Pahang, and he is also Founder and Advisor of the Innovative
Manufacturing, Mechatronics and Sports Laboratory. He teaches and conducts
research in the areas of industrial automation, robotics, ergonomics, sustainable
manufacturing, machine learning and sports engineering.
Prof. Junmo Kim received the B.S. from Seoul National University, Seoul, Korea,
in 1998, and the M.S. and Ph.D. from the Massachusetts Institute of Technology
(MIT), Cambridge, in 2000 and 2005, respectively. From 2005 to 2009, he was
with the Samsung Advanced Institute of Technology (SAIT), Korea, as Research
Staff Member. He joined the faculty of KAIST in 2009, where he is currently
Associate Professor of Electrical Engineering. His research interests are in image
processing, computer vision, statistical signal processing and information theory.
Longitudinal Velocity Control Design
with Error Tolerance Strategy
for Autonomous Vehicle
1 Introduction
The Fourth Industrial Revolution witnesses the birth of several emerging technologies
such as blockchain, ridesharing as well as the driverless vehicle. These technologies
possess highly disruptive effects towards the society [1]. Consequently, the innova-
tions have led to the discussion in several other interdisciplinary fields. For example,
“Society 5.0” is a term coined in Japan to describe the integration between the virtual
reality and physical space of the future society [2]. The discussions not only happen
in the highly developed countries in the developed region, but they also reached other
world regions, including South East Asia. Several high-profile investment corridors
have been launched in recent years with regards to the Autonomous Vehicle (AV)
field in Malaysia. For example, South Perak Region, a high-end economic corri-
dor is targeting to be the center of hardware production for AV components [3].
However, the real-time implementation of AV in Malaysia is still limited. Previous
works in Malaysia mostly focusing on the development of ADAS and Active Safety
implementation such as Collision Avoidance, among many others [4]. However, in
2018, the first driverless vehicle prototype has been developed in Malaysia with the
full SAE Level 5 automation. The first demo has been done in a closed environ-
ment to the public showcasing the prototype in the Malaysian Global Innovation &
Creativity Centre (MaGIC), Cyberjaya [5]. In this work, some of the preliminary
results during the development of the platform is denoted. It revolves around the
longitudinal velocity control strategy used in the platform. As a proof of concept
work, a series of experimental validation’s results are also included, to support the
controller’s feasibility. It is important to mention that since the objective of this work
is not to discuss the detailed technical aspects of the autonomous vehicle develop-
ment, thus, additional literature suggestions are included for interested readers in the
related topics.
The human driver controls the vehicle speed by providing alternate braking and
throttle actuation. For autonomous vehicle (AV), this requires a reliable longitudinal
motion control system. However, a velocity controller with high sensitivity to the
speed tracking error will produce an uncomfortable riding experience. This is due to
the controller tries to penalize the error in maximum effort and subsequently yielding
a rapid alternation between braking and throttle actuation. Despite this, particularly
for the low-speed urban ride (under 30 km/h), small speed tracking error does not
provide much difference. By having a too sensitive controller, the host vehicle will
jerk when minimizing the error due to rapid alternate actuators intervention. Thus,
the objective of this study is to develop a longitudinal velocity control design for
the autonomous vehicle with error tolerance strategy, to allow for a smooth speed
tracking performance by the AV.
Longitudinal Velocity Control Design with Error … 3
Fig. 2 Communication layer for the longitudinal motion control development of the prototype
The high-level controller acts to formulate the desired actuation percentage value
by the braking and gas pedals. It consists of three submodules, i.e. error tolerance
strategy, adaptive PID controller as well as the actuation selector. The inclusion of
error tolerance strategy prevents a too sensitive controller as well as to prevent abrupt
increment of controller’s output during initial navigation of AV, which subsequently
will hinder user’s comfort. By having some safe region relative to the error (Fig. 4),
the adaptive PID penalizes the error with an adaptive gain formulation. The selector
Longitudinal Velocity Control Design with Error … 5
Fig. 3 Longitudinal velocity control design for the driverless prototype platform
Fig. 4 Error tolerance strategy used for the adaptive PID controller formulation and tuning
then acts as the safety feature to prevent simultaneous gas and throttle actuation,
which is infeasible in real navigation.
As shown in Fig. 4, the error tolerance strategy acts by providing a constrained
region for the real-time tracking error between the desired and current velocity value.
The conditioning is based on the online information from the tracking error. By
providing the maximum and minimum allowed error, for the lower speed navigation
(under 30 km/h), this will prevent a too sensitive controller action. In addition, it
will also aid the adaptive tuning process of the PID strategy. The control output is
obtained by tracking the error based on the proportional, integral and derivative terms
6 U. Z. A. Hamid et al.
(P, I and D respectively). The nominal formulation for the control function, u(t) for
the PID are as follows:
t
de(t)
u(t) = K p e(t) + K i e t dt + K d (1)
dt
0
where K p , K i and K d are respectively the gain for P, I and D, and e represents the
tracking error. More details on the formulation of the adaptive PID controller is
omitted due to limited space in this work and can be found here [7, 8]. Due to the low-
speed implementation of the prototype, the usage of PID is sufficient for the high-level
controller due to the absence of high nonlinearity of the vehicle dynamics. However,
future works of more complex navigation scenario demand the implementation of
optimal controller usage.
4 Experimental Design
To test the ability of the proposed design in providing smooth alternate braking
and throttle intervention, the controller has been tested on a real AV platform in a
series of varied low-speed controlled environment scenario (30 km/h). Due to the
prototype is aimed to performed in the controlled environment, the test is done in a
straight road testing area nearby Universiti Teknologi Malaysia, Kuala Lumpur. The
host vehicle initially is in a non-moving state. It starts to navigate autonomously by
tracking the desired velocity, as depicted in Fig. 3. The prototype is validated in two
scenarios. The objective of Scenario 1 is to show the ability of the velocity tracking
by the autonomous platform during constant navigation, where the speeds are rarely
changing. In addition, the vehicle is required to prevent from providing sudden high
input to the throttle. For Scenario 2, in a real autonomous navigation, particularly in
the obstacle’s existence, the reference generator usually provides constantly changing
value of the reference velocity. A too sensitive controller will yield jerky navigation.
Thus, Scenario 2 is based on this requirement. Figure 5 illustrates the used platform
and Fig. 6 depicts the straight road location of the testing. The vehicle navigates from
point red to point red, with different reference velocity. A safety driver is present
during the testing session. The description of both proposed test scenarios is discussed
in their respective subsection.
Scenario 1 The aim of this scenario is to validate the inclusion of error tolerance
strategy into the controller in yielding and prevent too sensitive control output reac-
tion. The vehicle initially starts to navigate autonomously in a straight road with
the reference velocity of 10 km/h. The velocity reference is then changed to 20 and
30 km/h intervally after several moments of navigation.
Scenario 2 The aim of this scenario is to validate the inclusion of error tolerance
strategy into the controller in yielding and prevent too sensitive control output reac-
Longitudinal Velocity Control Design with Error … 7
Fig. 5 The prototype, a collaboration between Moovita and Universiti Teknologi Malaysia
tion in the event when the desired velocity is rapidly changing. The maximum speed
is 30 km/h. This usually happens in the occurrence of risk, hazard or traffic jam.
Thus, to ensure comfort, error tolerance strategy is expected to prevent too sensitive
performance as well as hindering the simultaneous gas and braking actuation.
For Scenario 1, as can be seen in Fig. 7, the host vehicle is able to track the desired
velocity. The inclusion of error tolerance strategy allows for alternate actuation of
the vehicle and prevents sudden increment of control input (gas) during the initial
navigation. While for Scenario 2 in Fig. 8, the proposed algorithm formulation is
shown to aid the vehicle to navigate smoothly during the rapid change of the desired
velocity. As can be seen, the gas and braking are yielded alternately, and the error
tolerance strategy prevents too rapid actuation alternation between the gas and brak-
ing in maintaining the desired velocity. This show that the proposed controller is
feasible for a real-time implementation and this work serves its purpose as the proof
of concept paper.
Overall, from the experimental results, the vehicle navigates with smooth alternate
braking and throttle actuation in both scenarios. The inclusion of error tolerance
strategy increased the riding comfort and prevent vehicle jerking experience due to
the rapid alternation between braking and throttle intervention. The control design
allows soft violation of the error and at the same time ensuring the error is not
large enough by enforcing tracking error constraints. For example, in Fig. 8a, for the
8 U. Z. A. Hamid et al.
Fig. 6 Top Google Maps view of the experimental validation location. The host vehicle navigates
from red point to another red point with varied reference velocity
Fig. 7 Results for scenario 1, where the velocity tracking, tracking error, controller adaptive gain
as well as the braking & throttle percentage during the navigation is shown
with regards to the vehicle dynamics and longitudinal motion performance such as
jerking can be added in the future.
6 Conclusion
A longitudinal motion controller design for AV has been proposed. The inclusion of
error tolerance strategy allows for the soft violation of the error penalization. This
prevents the AV from jerking scenario due to the rapid alternation between braking
and throttle during the tracking performance. This in return enables a smooth and
reliable navigation. This will subsequently allow it to be implemented into AV’s
usage in the low-speed urban area. In addition, this work is important for the future
development of AV works in the South East Asian region (Singapore, Malaysia).
For future works, efforts must be done to have a proper experimental validation area
for more complex scenario implementation. In addition, works on the passenger’s
comfort should be pursued to increase the user’s acceptance to the driverless vehicle
technology.
10 U. Z. A. Hamid et al.
Fig. 8 Results for scenario 2, where the velocity tracking, tracking error, controller adaptive gain
as well as the braking & throttle percentage during the navigation is shown
References
1. Rahman AA, Hamid UZA, Chin TA (2017) Emerging technologies with disruptive effects: a
review. Perintis E-J 7(2):111–128
2. Koezuka H (2016) Utilization of IoT in the long-term care field in Japan. In: 2016 2nd inter-
national conference on cloud computing and internet of things (CCIOT). IEEE, New York, pp
166–169
3. Tanjung Malim, South East Asia’s Next Generation Automotive Centre, South Perak
Region http://www.southperak.com.my/4-nodes/node-2-tanjung-malim-south-east-asias-next-
generation-automotive-hub. Accessed 03 May 2018
4. Hamid UZA, Zakuan FRA, Zulkepli KA, Azmi MZ, Zamzuri H, Rahman MAA, Zakaria MA
(2018) Multi-actuators vehicle collision avoidance system-experimental validation. In: IOP con-
ference series: materials science and engineering, vol 342, No 1. IOP Publishing, p 012018
5. Universiti Teknologi Malaysia leading the Autonomous Driving R&D field resulting to
1st Demo in Malaysia https://news.utm.my/2018/01/universiti-teknologi-malaysia-leading-the-
autonomous-driving-rd-field-resulting-to-1st-demo-in-malaysia/. Accessed 03 May 2018
Longitudinal Velocity Control Design with Error … 11
6. Taş ÖŞ, Kuhnt F, Zöllner JM, Stiller C (2016) Functional system architectures towards fully
automated driving. In: 2016 IEEE intelligent vehicles symposium (IV). IEEE, New York, pp
304–309
7. Ang KH, Chong G, Li Y (2005) PID control system analysis, design, and technology. IEEE
Trans Control Syst Technol 13(4):559–576
8. O’Dwyer A (2009) Handbook of PI and PID controller tuning rules. Imperial college press,
London
Vision Based Smart Sorting Machine