发展论坛 (Development Forums)
Development Forum 4:
7月27日13:30-17:00
July 27, 13:30-17:00
SDGs, Mobility, and System Control
Chair: Tielong Shen (Sophia University, Japan)
Panelists: Toru Namerikawa, Keio University, Yokohama, Japan
Yuji Yasui, Honda R&D Ltd, Japan
Kunihiko Suzuki, Hitachi Astemo, Ltd., Japan
Yuhu Wu, Dalian Institute of Technology, China
Zhenhui Xu, Sophia University, Tokyo, Japan
Abstract:The Sustainable Development Goals (SGDs) have been “a shared blueprint for peace and prosperity for people and the planet.” Related to the prosperity, social mobility is still under growing fields since mobility of people and transport of goods are engines of developing. However, the rapidly growing mobility have caused global environment issues. To solve these issues, innovation of automotive technology such as connected and automated vehicles, electrification of vehicle powertrains, and revolution in social mobility such as mobility as a service (MaaS) have proposed in the last few years, which are strongly related to the system and control sciences.
This CCC2022 development forum will focus on the system and control sciences for the mobility. It will discuss what the essential problems in improving energy efficiency and reducing CO2 emission have to challenge from the view of system and control theory. Furthermore, new challenges from automotive industry will be introduced which will show a state-of-art of technology innovation. The speakers are invited from both of academia and industrial community.
Agenda:
13:30-13:50 Prologue: Mobility, Social Optimization, and Large-population Control
Prof. Tielong Shen, Sophia University, Tokyo, Japan
13:50-14:40 Keynote I: Resilient and Secure Estimation for Cyber-Physical Systems
Prof. Toru Namerika, Keio University, Yokohama, Japan
14:40-15:10 Reinforcement Learning and Mean Field Game
Dr. Zhenhui Xu, Sophia University, Tokyo, Japan
15:10-15:25 Coffee Break
15:25-16:05 Keynote II: Honda’s Technical Strategy toward Carbon Neutrality and Freedom of
Mobility in Future
Dr. Yuji Yasui, Honda R&D Ltd., Japan
16:05-16:45 Keynote III: . Big Data Analytics for More Sophisticated Powertrain MBD
Dr. Kunihiko Suzuki, Hitachi Astemo, Ltd., Japan
16:45-17:15 The Vehicle Routing Problem with Charging Relief in Post-Disaster
Prof. Yuhu Wu, Dalian Institute of Technology, China
Speakers’ Abstract and Biography
Title: Mobility, Social Optimization, and Large-population Control
Abstract: The aim of this forum will be explained with a short introduction of motivated background. Challenging issues in mobility system and control theory will be addressed by giving a couple of examples including large-scale of electric vehicles and traffic flow. The issues will be further addressed by the following speakers.
Tielong Shen is a Full Professor in control engineering at Sophia University, Tokyo, Japan. His research interests include control theory and applications in automotive powertrain systems, power systems, and mechanical systems. Dr. Shen has author/co-authored eleven text books in Japanese, English and Chinese, respectively, and has published more than 200 research papers in major journals. He has served SICE, TCCT of CAA, IFAC and IEEE as Chair/vice-chair including General Chair of SICE&CCC2015, IPC Chair of IFAC AAC2016 etc. The last year, he served as general chair of SICE Annual Conference 2021, and general chair of IFAC Conference on ECOSM 2021. Recently, he is awarded the 8th TCCT Outstanding Contribution Award.
Title:Resilient and Secure Estimation for Cyber-Physical Systems
Abstract: Advances in networks and information technologies have great potential for the building of smart societies. These technologies contribute to enhancing sustainability and efficiency of social systems by interacting with physical infrastructures, and such systems that integrate physical entities and cyber components are referred to as cyber-physical systems (CPS). Many systems, such as energy, transportation, medical, and manufacturing systems, are considered as CPS, and along with the innovation of Internet of things (IoT), the importance of the systems will further increase. However, CPS are severely dependent on computing and networking elements, and hence CPS have several opportunities for malicious third parties to inject attacks. According to the solid interaction between cyber and physical space, the adversaries’ action against CPS conduces tragic consequences to physical entities, not only cyber components. Therefore, for the sake of secure operations of CPS, we need to consider the cyber threat scenarios and strengthen the security and resilience of CPS.
This talk deals with the problem of secure state estimation in an adversarial environment with the presence of noises. We show that the prior information of the estimated state enhances the system resilience, that is, even if more sensors are compromised, when the information of the state is given, then one can uniquely recover the state and attack. The problem can be represented as a min-max optimization, that is, the system operator seeks an optimal estimate which minimizes the worst-case estimation error due to the manipulation by the attacker. Some numerical comparisons and examples finally illustrate the effectiveness of the proposed estimators.
Toru Namerikawa received the B.E., M.E., and Ph.D. degrees in electrical and computer engineering from Kanazawa University, Kanazawa, Japan, in 1991, 1993, and 1997, respectively. From 1994 to 2002, he was an Assistant Professor with Kanazawa University. From 2002 to 2005, he was an Associate Professor with Nagaoka University of Technology, Niigata, Japan. From 2006 to 2009, he was again with Kanazawa University. In April 2009, he joined Keio University, Yokohama, Japan, where he is currently a Professor with the Department of System Design Engineering.
He has held visiting positions at Swiss Federal Institute of Technology in Zurich in 1998, University of California, Santa Barbara in 2001, University of Stuttgart in 2008 and Lund University in 2010.
He received the 2014 Pioneer Technology Award from SICE Control Division and the 2017 Outstanding Paper Award from SICE.
He is an Editor of SICE Journal of Control, Measurement, and System Integration, an Associate Editor of IET Control Theory & Applications.
His main research interests are robust control, distributed and cooperative control, and their application to cyber-physical systems including power network and transportation network systems. He is a senior member of IEEE.
Title: Reinforcement Learning and Mean Field Game
Abstract: Mean field game (MFG) is a new branch of game theory. It designated a greatly effective methodology for analyzing differential games with a large number of interacting agents. This analytical machinery sheds light on the formation of collective behavior in large complex decision models involving many agents by relating it to the micro-behavior of individuals. The overall methodology provides the ground to develop rich results in stochastic analysis, control theory, partial differential equations, and numerical analysis. Meanwhile, with the development of reinforcement learning technology and availability of cheaper measuring and computing equipment, many data-based strategy design methods have been developed in the control systems community. This talk will present a new data-based approximate Nash equilibria design method for mean field games.
Zhenhui Xu received the M.S. degree in control engineering from University of Science and Technology of China, in 2017 and the Ph.D. degree from mechanical engineering, Sophia University, Tokyo, Japan, in 2021. Since 2021, she has been a Postdoctoral Fellow with the Department of Engineering and Applied Sciences, Sophia University. Her research interests are in the fields of learning-based optimal control and nonlinear systems.
Title:
Honda’s Technical Strategy toward Carbon Neutrality and Freedom of Mobility in Future
Abstract: We are facing various social issues such as global warming, energy shortage, labor shortage and traffic accidents. However, Honda is confident of that we can resolve these issues by evolution of technology. The powertrains of automobiles will be changed from simple internal-combustion engines to electrical motors supported by next-generation batteries, fuel-cell, etc. and to hybrid concept ones. Highly-intelligence vehicles and robots will support freedom of mobility of humans and goods. Social services such as MaaS (Mobility as a Service) also will be very important. Consequently, Honda set up the 2030 vision of “Serve people worldwide with the joy of expanding their life’s potential” and are researching and developing various technologies for mobility products and services. Honda’s technical strategy toward carbon neutrality and freedom of mobility and latest control technologies for mobility products and services for future will be introduced in this speech.
Yuji Yasui received the B.E. and M.E. degrees in mechanical engineering from
Tokyo University of Science, Japan, in 1992 and in 1994, and the Ph.D degree from Sophia University, Japan, in 2012. He joined Honda R&D Co., Ltd in 1994 and has been researched powertrain control for low-emission vehicles and HEVs, traction control for F-1 racing car, transmission control and device control by using adaptive control, model predictive control, neural network, machine learning, etc. He moved to research area for automated driving and driving support systems using AI and advanced control technologies in 2016. He is currently an executive chief engineer in a research group for intelligent technology for various mobility.
Title: Big Data Analytics for More Sophisticated Powertrain MBD
Abstract: To comply with strengthening environmental regulations, the control software becomes larger, as the powertrain system complexity increases. Since the control software development cost and workload drastically increase due to the system functionality enhancement, more efficient control design process and methodologies are essentially required. To address these issues, auto-industry has proactively evolved MBD, which allows for efficient rapid prototyping and advanced validation utilizing physical simulation techniques for among component, system level design and testing, for last two decades.Towards connected-cars era based on IoT platform, more sophisticated MBD process would be achieved through fusion of physics-based and data-driven approaches, such as, machine learning, neural network surrogate modeling, and data assimilation etc. In the context of these circumstances, we also tackle the challenges with the existing 3D/1D CAE combining with the data-driven engineering. The status and prospects of our powertrain MBD process will be described through the introduction of our several case studies.
Kunihiko Suzuki joined Hitachi Research Laboratory of Hitachi, Ltd., Japan in 2003. He is in charged with R&D of automotive gasoline engine management systems with auto-manufactures and led the establishment of powertrain MBD environment in the firm. He received the Ph.D. degrees in mechanical engineering from Ritsumeikan University, Japan in 2005. In 2013, he was sent to Hitachi (China) R&D Corporation in Shanghai to coordinate collaborative projects with Chinese universities and auto-manufactures. In 2015, he returned to Japan and was Unit Leader of corporate R&D group of Hitachi, Ltd. In 2021, He was transferred to Technology Development Division of Hitachi Astemo, Ltd. Currently, he is a manager of Powertrain Technology Development Department. His interests are in intelligent control, big data analytics, their powertrain control application, and industrialization.
Title: The Vehicle Routing Problem with Charging Relief in Post-Disaster
Abstract: When disasters cause damages to power systems, people’s daily life will be affected. This talk will focus on emergency transportation for power recovery in post-disaster and formulates the problem as a mixed-integer linear programming model, which called vehicle routing problem with charging relief (VRPCR). As we know, the state of charge (SoC) is the available capacity of a battery as a percentage of its total capacity, which implies the working hours that the battery can provide. Our goal is to make a set of shelters charge before the shelter battery SoC reaches the minimal value over time. Furthermore, the related time constraints of each shelter can be calculated from the initial value and the evolution of battery SoC at each shelter. Since the large scale and complex road networks will be generated in post-disaster, we develop a two-stage algorithm to deal with the problem. To this end, a reduced road network is received from a leading road network in stage 1 by A-star algorithm. Subsequently, to determines the delivery order of shelters following the related problem constraints and objective function by enhanced genetic algorithm in stage 2. Simulation results clearly demonstrate that our method is able to achieve satisfactory solutions.
Yuhu Wu received the PhD degree in Mathematics from Harbin Institute of Technology, Harbin, China, in January 2012. Since September 2012, he has held an assistant professor position at Harbin University of Science and Technology, China. He held a postdoctoral research position at Sophia University, Japan, from April 2012 to September 2015. In October 2015, he joined the School of Control Science and Engineering, Dalian University of Technology, China, where he is currently a professor. His research interests are related to non-linear control theory, game theoretic control, and applications of control to automotive powertrain systems, and unmanned aerial vehicles. He is serving as Associate Editor for Asian Journal of Control.
