Learning-based decision and control systems for high-level self-driving intelligence
	Keqiang Li
	Professor, Tsinghua University, Beijing, China Academician, Chinese Academy of Engineering, China

Abstract:

Today’s autonomous driving system are facing severe challenges in highly dynamic, random and dense traffic scenarios. Existing hierarchical design method, for example, that with rule-based decision and linear motion controller, is lack of sufficient adaptability and flexibility. As a biologically inspired artificial intelligence, reinforcement learning (RL) is promising to provide the self-evolving ability for automated cars, which has the potential to generalize to unknown driving scenarios. This talk will discuss recent advances in learning-based autonomous driving systems for high-level self-driving intelligence. An interpretable and computationally efficient framework, called integrated decision and control (IDC), is proposed to fulfill more flexible functionality, in which the standard actor-critic architecture can be subtly utilized to train its decision and control neural networks. Some technical breakthroughs in safe reinforcement learning and high-fidelity simulator are also discussed for the purpose of training more accurate neural network controllers.

Bio:

Dr. Keqiang Li is currently a professor at School of Vehicle and Mobility, Tsinghua University. He is the Academician of Chinese Academy of Engineering. He also serves as the director of State Key Laboratory of Intelligent Green Vehicle and Mobility, and chief scientist of National Innovation Center of Intelligent and Connected Vehicles. Dr. Li is an expert in the field of automotive intelligence. His main research areas include dynamic design and intelligent control of driver assistance systems / autonomous driving systems. He has authored about 250 journal/conference papers and over 80 patents in and outside of China. He has worked in Japanese and Germany automotive companies and academic institutions for many years including Tokyo University of Agriculture and Technology, University of Tokyo, Aachen University of technology, National Traffic Safety & Environment Lab in Japan, Isuzu Automobile Corp, etc. Dr. Li is also the recipient of Changjiang Scholar Program Professor, China National Technological Invention Award, and China National Scientific and Technological Progress Award.

---

	Improving the efficiency of combustion engines with combustion control
	Per Tunestål
	Professor, Lund University, Sweden

Abstract:

Combustion engines powered by fossil fuels are large contributors to greenhouse gas emissions. Introduction of renewable fuels, both as drop-in and neat fuels, can however reduce these emissions quite substantially and make the combustion engine part of the solution rather than the problem. Even with renewable fuels however, we need to economize with our resources and keep improving the efficiency of combustion engines. Combustion control has an important role here making sure the engines perform as intended not just on the test bench but also in reality and throughout their service life. This presentation presents a range of combustion control solutions developed at Lund University over many years ranging from supervisory control of conventional compression ignition engines using midranging to cycle-to-cycle and even in-cycle model predictive control of advanced combustion concept.

Bio:

Per Tunestal received his PhD in Mechanical Engineering at the University of California, Berkeley in 2000. He presently holds a position as Professor at Lund University where he is in charge of the engine control activities. Per Tunestal has previously served as Director of The KCFP Engine Research Center, a consortium financed by The Swedish Energy Agency, Lund University and 14 member companies world-wide. Special interests are engine control based on in-cylinder measurements and cylinder-pressure based parameter estimation. Per Tunestal holds more than 130 scientific publications within the combustion engine field and he has served as chairman of the Control and Calibration committee within the Society of Automotive Engineers. Lund university was founded in 1666. Today it is an international center for research and education that has approximately 48 000 students and 7500 employees. Lund University is respected as one of the top universities in Sweden with an excellent academic reputation and a large number of visiting professors and international students. Lund University is also consistently ranked as one of the top 100 universities in the world.

---

	Advances in MPC for Real-Time Applications in Connected Automotive Systems
	Hong Chen
	Professor, Tongji University, China

Abstract:

Nonlinear model predictive control (NMPC) is very suitable for practical industrial applications but subjected to the problem of computational burden. This presentation will introduce the latest technologies in fast solution of NMPC and its application to connected automotive systems from the perspective of predictive energy control system, including predictive energy-Saving control, energy-heat collaborative predictive control and multi-vehicle collaborative predictive control for high efficient transportation.

Bio:

Hong Chen is an IEEE Fellow, CAA-China Fellow and SAE-China Fellow. She received the B.S. and M.S. degrees in process control from Zhejiang University, China, in 1983 and 1986, respectively, and the Ph.D. degree in system dynamics and control engineering from the University of Stuttgart, Germany, in 1997. In 1986, she joined Jilin University of Technology, China. From 1993 to 1997, she was a Wissenschaftlicher Mitarbeiter with the Institut fuer Systemdynamik und Regelungstechnik, University of Stuttgart. Since 1999, she has been a professor in Jilin University and hereafter a Tang Aoqing professor. Since 2019, she has worked at Tongji University as a distinguished professor and currently serves as Dean of the College of Electronic and Information Engineering. Her current research interests include model predictive control, automotive control and automated driving.

 

---

	Visual Sensing and Perception for Autonomous Driving
	Kuk-Jin Yoon
	Professor, Department of Mechanical Engineering, KAIST, Korea

Abstract:

Visual sensing and perception are fundamental building blocks for achieving safe and reliable autonomous driving. This talk will explore the critical role of visual sensors such as cameras, Lidar, and Radar, and cutting-edge computer vision techniques in enabling autonomous vehicles to understand their surroundings. I will discuss various deep learning algorithms used to extract meaningful information from visual data and also address challenges and limitations of visual perception in autonomous driving. Finally, I will explore emerging trends and future directions in this rapidly evolving field.

Bio:

Kuk-Jin Yoon is a professor in the Department of Mechanical Engineering at the Korea Advanced Institute of Science and Technology (KAIST), where he leads the Visual Intelligence Laboratory. He earned his BS, MS, and Ph.D. degrees in electrical engineering and computer science from KAIST in 1998, 2000, and 2006, respectively. From 2006 to 2008, he was a postdoctoral fellow with the PERCEPTION Team at INRIA in Grenoble, France. Subsequently, he served as an assistant/associate professor at the School of Electrical Engineering and Computer Science at the Gwangju Institute of Science and Technology (GIST) in South Korea from 2008 to 2018. His research focuses on computer vision techniques for autonomous driving.

---

	Engine control with specific drop-in fuel to reduce CO2 emission
	Yasuo Moriyoshi
	Professor, Chiba University, Japan Head, Engine Research Center, Chiba University

Abstract:

It has been found that by reforming gasoline to increase the olefin component with high octane sensitivity, the lean and EGR-diluted combustion limits can be extended. It has been also found that if engine control that makes maximum use of these characteristics can be realized, CO2 emissions can be reduced by up to 10% in WLTC mode driving. This presentation introduces how to effectively reduce CO2 emission by controlling different base engine specifications and base vehicles (gasoline engine, supercharged gasoline engine, series hybrid engine, and parallel hybrid engine).

Bio:

Dr. Yasuo Moriyoshi obtained his PhD from Tokyo Institute of Technology in 1990 and joined Chiba University. He is now a full professor and also the head of Center for Power Source Research for Next-Generation Mobility at Chiba University.

He has been engaged in internal combustion engine research, such as in-cylinder gas motion, fuel spray, stratified charge combustion, cycle-to-cycle variation, HCCI, combustion control, ignition system, abnormal combustion and boosting system. Also, his current interest is real driving emissions and traffic flow control.

He has established a venture company of "SERC" in 2011 to collaborate engine research, such as consortia on highly boosted gasoline engines and co-generation natural gas engines.

He is the fellow of SAE, JSAE and JSME.

 

---

	Research on Optimization of Vehicle Energy Management for NEV
	Feng Ding
	Senior Technical Director, United Automotive Electronic Systems Co., Ltd.

Abstract:

NEV performance would be limited at low environment temperature, electric range shortening and other issues are the focus of OEM and drivers. NEV requires higher overall system efficiency during full temperature range. And the application of high efficiency efficient thermal management system and connected technology on NEV brings more possibilities to the optimization of vehicle energy management. Therefore, Vehicle Energy Management (VEM) proposed to coordinate control of the internal combustion engine, electric drive and thermal management system in vehicle level, focusing on efficiency, performance and user experience improving. The research of VEM tries to optimize from three aspects: scenario, efficiency and multi-objective coordination optimization. Application scenario analysis, vehicle energy flow testing and analysis have been done. Based on the analysis, considering driving and thermal needs, vehicle energy management solution is formed and advanced functions for multiple scenarios have been developed.

 

 

Bio: Feng Ding graduated from Beijing University of Aeronautics and Astronautics with a Master's degree in Automatic Control. Responsible for cross domain controller development under latest EEA and SOA in Intelligent connected vehicles. Rich professional knowledge and experience in the design of automotive electronic controller, including system, software and hardware architecture, familiar with the current and next generation technical status, including vehicle energy, motion and vehicle-cloud integration control technologies.
He has won multiple awards for technological progress in the automotive industry and Shanghai, and was honored with titles such as Excellent Engineering Technology Leader by SAIC in 2019.