Focused Track 3:
Automotive Control
This is one of the three Focused Tracks of ICCAS 2020, and covers the following topics:
- Connected and Automated Vehicles
- AI/ML Applications for ADAS
- Security and safety for ADAS/AV
- Mapping/Localization
- Navigation and Motion Planning
- Perception for ADAS/AV
- Driver-Vehicle Interaction
- Multi-Vehicle Cooperative Automation
- Intelligent Transportation Systems
- Fault tolerant control design for AV
- Powertrain, Electrification, and Vehicle Dynamic Controls
- Engine/Powertrain/Aftertreatment Systems
- Alternative Fuels/Advanced Combustion Technologies
- Energy Conversion, Waste Heat Recovery
- Drivetrain/Driveline Systems, Vehicle Dynamics
- Steering and Suspensions Systems
- Hybrid and Electric Vehicles
- Battery and Energy Storage Systems
- Vehicle Active and Passive Safety Systems
Plenary Speaker
Anna Stefanopoulou, University of Michigan, Ann Arbor, USA
Invited Speakers
- Shengbo Li, Tsinghua University, China
- Pongsathorn Raksincharoensak, Tokyo University of Agriculture and Technology, Japan
 |
Distributional Soft Actor Critic (DSAC) and Its Application on Autonomous Driving
Prof. Shengbo Eben LI |
| Abstract: Reinforcement learning (RL) has been successfully applied to a range of challenging sequential decision making and control tasks such as games and robotics. However, current RL algorithms typically suffer from the Q-value overestimation problem, which will greatly reduce policy performance and limit the applicability of RL to real-world domains. This talk will introduce the distributional soft actor-critic (DSAC) algorithm, which is recently developed for mitigating Q-value overestimations. The main theoretical basis of DSAC is that learning a distribution function of state-action returns can effectively mitigate Q-value overestimations because it is capable of adaptively adjusting the update step of the Q-value function, thereby improving policy performance. Besides, the application of DSAC in decision making and motion control of autonomous vehicles will also be introduced.
Bio: Dr. Shengbo Li received the M.S. and Ph.D. degrees from Tsinghua University in 2006 and 2009. Before joining Tsinghua University, he has worked at Stanford University, University of Michigan, and UC Berkeley. He is now leading Intelligent Driving Lab (iDLab) at Tsinghua University. His active research interests include intelligent vehicles and driver assistance, reinforcement learning and optimal control, distributed control and estimation, etc. He is the author of over 100 peer-reviewed journal/conference papers, and the co-inventor of over 30 patents. Dr. Li was the recipient of Best Paper Award in 2014 IEEE ITS, Best Paper Award in 14th Asian ITS. He also serves as Board of Governor of IEEE ITS Society, AEs of IEEE ITSM, IEEE Trans ITS, etc. |
|
 |
Risk Predictive Driver Assistance System towards Zero-Traffic Accidents
Prof. Pongsathorn Raksincharoensak |
| Abstract: This talk describes a risk predictive driver assistance system which is designed to prevent potential crashes in early stage while still keeping human drivers in the control loop. The proposed system focuses on two key technologies: Risk-predictive driving characteristics and Haptic shared control interface between the driver and the assistance system. Risk prediction is modeled based on analysis on near-crash incident database of Smart Mobility Research Center of Tokyo University of Agriculture and Technology. Especially, the risk predictive scenario focuses on the occlusion in the road where pedestrians or cyclists might suddenly appear. Experiments using the test vehicle are conducted to verify the effectiveness and investigate the driver acceptance.
Bio: He is currently a full professor of department of Mechanical Systems Engineering at Tokyo University of Agriculture and Technology. His Research interests include vehicle dynamics and control in the context of active safety, handling dynamics, collision avoidance and driver behavior modeling. He has received a number of awards such as Society of Automotive Engineers of Japan(JSAE) Excellent Journal Paper Award (2018), JSAE Asahara Scientific Award (2014), Masao Horiba Award for “Autonomous Driving Technology” (2016), Best Paper Awards in International Symposia : AVEC, FAST-zero, APAC, FISITA. He is currently a general secretariat of International Symposium of Advanced Vehicle Control (AVEC) Board. Recently, he is one of co-authors of Vehicle Dynamics of Modern Passenger Cars, published by Springer. |
|
Organizing Committee Chair
Dongsuk Kum, KAIST, Korea
Organizing Committee
- Shengbo Li, Tsinghua University, China
- Pongsathorn Raksincharoensak, Tokyo University of Agriculture and Technology, Japan
- Youngjin Park: KAIST, Korea
- Kyoungsoo Kim: KAIST, Korea
- Chung Choo Chung: Hanyang University, Korea
- Seibum Choi: KAIST, Korea
- Kyoungdae Kim: DGIST, Korea
- Seangwock Lee: Kookmin University, Korea
- Yeonsik Kang: Kookmin University, Korea
- Youngbae Hwang: Chungbook National University, Korea
- Bongseob Song: Ajou University, Korea
- Kibeom Lee: Halla University, Korea
Program
Oct. 15 (Thursday) 16:30~18:00 Online Room 6
| TC6.1 | 16:30~16:45 | Recurrent Neural Network to Estimate Intake Manifold O2 Concentration in a Diesel Engine (62) | Loris Ventura (Politecnico di Torino)*; Stefano Malan (Politecnico di Torino) |
| TC6.2 | 16:45~17:00 | NLQR Control of High Pressure EGR in Diesel Engine (81) | Loris Ventura (Politecnico di Torino)*; Stefano Malan (Politecnico di Torino) |
| TC6.3 | 17:00~17:15 | Development of Steering Control Algorithms with Self-tuning Fuzzy PID for All-terrain Cranes (110) | Jaho Seo (Ontario Tech University)*; Moohyun Cha (Korea Institute of Machinery & Materials); Kwangseok Oh (Hankyong National University); Young-Jun Park (Seoul National University); Tae J. Kwon (University of Alberta) |
| TC6.4 | 17:15~17:30 | Model Predictive Path Planning Based on Artificial Potential Field and Its Application to Autonomous Lane Change (123) | Pengfei Lin (Hanyang University); Woo Young Choi (Hanyang University); Seung-Hi Lee (Hanyang University); Chung Choo Chung (Hanyang University)* |
| TC6.5 | 17:30~17:45 | Predictive Collision Avoidance Control with Optimized Ride Comfort in Vehicle Lateral Motion Control (147) | Jin Ho Yang (Hanyang University); Dae Jung Kim (Hanyang University); Chung Choo Chung (Hanyang University)* |
| TC6.6 | 17:45~18:00 | Clutch Torque Estimation of Ball-ramp Dual Clutch Transmission using Higher Order Disturbance Observer (183) | Dong-Hyun Kim (KAIST)*; Seibum Choi (KAIST) |
| TC6.7 | 18:00~18:15 | Nonlinear Model Predictive Control for Self-Driving cars Trajectory Tracking in GNSS-denied environments (204) | Ali Barzegar (Kunsan National University)*; Oualid Doukhi (Kunsan National University); Deok-Jin Lee (Kunsan National University); Yeon-ho Jo (Kunsan National University) |
Oct. 16 (Friday) 10:30~12:00 Online Room 5
| FA5.1 | 10:30~11:00
Keynote |
Distributional Soft Actor Critic (DSAC) and Its Application on Autonomous Driving | Prof. Shengbo Li (Tsinghua University) |
| FA5.2 | 11:00~11:15 | Decision of Driver Intention of a Surrounding Vehicle Using Hidden Markov Model with Optimizing Parameter Estimation | Jin Ho Yang (Hanyang University); Dae Jung Kim (Hanyang University); Tae Won Kang (Hanyang University); Jeong Sik Kim (Hanyang University); Chung Choo Chung (Hanyang University)* |
| FA5.3 | 11:15~11:30 | Autonomous Evasive Steering with Differential Braking Backup | Moad Kissai (ENSTA Paris)*; Anh-Lam Do (Renault); Xavier Mouton (Renault); Bruno MONSUEZ (ENSTA ParisTech) |
| FA5.4 | 11:30~11:45 | Longitudinal and Lateral Integrated Safe Trajectory Planning of Autonomous Vehicle via Friction Limit | Kibeom Lee (Halla University); Dongsuk Kum (Korea Advanced Institute of Science and Technology)* |
| FA5.5 | 11:45~12:00 | Finite State Machine based Vehicle System for Autonomous Driving in Urban Environments | SangHyeon Bae (Sungkyunkwan University)*; Sung-Hyeon Joo (SungKyunKwan University); Jung-Won Pyo (SungKyunKwan University); Jae-Seong Yoon (SungKyunKwan University); Taeyong Kuc (SungKyunKwan University ); GwangHee Lee (Korea Institute of Industrial Technology) |
Oct. 16 (Friday) 13:20~14:50 Online Room 5
| FB5.1 | 13:20~13:50
Keynote |
Risk Predictive Driver Assistance System towards Zero-Traffic Accidents | Prof. Pongsathorn Raksincharoensak (Tokyo University of Agriculture and Technology) |
| FB5.2 | 13:50~14:05 | Mixed Reinforcement Learning for Efficient Policy Optimizationin Stochastic Environments | Yao Mu (Tsinghua University); Baiyu Peng (Tsinghua University); Ziqing Gu (Tsinghua University); Shengbo Li (Tsinghua University)*; Chang Liu (Cornell University); Bingbing Nie (Tsinghua University); Jianfeng Zheng (Didi Chuxing); Bo Zhang (Didi Chuxing) |
| FB5.3 | 14:05~14:20 | IMM EKF based Sensor Fusion for Vehicle Positioning Under Various Road Surface Conditions | Hyeon Uk Heo (Hanyang University); Dae Jung Kim (Hanyang University); Chung Choo Chung (Hanyang University)* |
| FB5.4 | 14:20~14:35 | 3D SaccadeNet: A Single-Shot 3D Object Detector for LiDAR Point Clouds | Lihua Wen (University of Ulsan)* |
| FB5.5 | 14:35~14:50 | Deep Reinforcement Learning-based ROS-Controlled RC Car for Autonomous Path Exploration in the Unknown Environment | Sabir Hossain (Kunsan National University); Oualid Doukhi (Kunsan National University); Yeon-ho Jo (Kunsan National University); Deok-Jin Lee (Kunsan National University)* |
