Lane Changing of Autonomous Vehicles in Mixed Traffic Environments: A Reinforcement Learning Approach

The emergence of connected and autonomous vehicles (CAVs) presents increased opportunities to mitigate traffic congestion, improve safety and reduce accidents. Professor Zhong-Ping Jiang, and researchers Leilei Cui and Sayan Chakraborty are applying innovative reinforcement learning control methods to one challenging aspect of CAV control: lane changing in mixed traffic. The team takes a novel approach by reducing the trajectory planning and tracking problem down to the minimization of a cost function that depends on a target way-point in the lane a CAV is targeting. They’ll discuss the integration of reinforcement learning and adaptive/approximate dynamic programming methods without assuming exact knowledge of surrounding vehicles, while avoiding the curses of dimensionality and modeling of conventional dynamic programming, and they’ll share simulation and validation results of this promising method towards minimizing fuel consumption and improve safety of the whole traffic stream.

Presenters

Professor Zhong-Ping Jiang is known for his contributions to stability and control of interconnected nonlinear systems, and is a key contributor to the nonlinear small-gain theory. His recent research focuses on robust adaptive dynamic programming, learning-based optimal control, nonlinear control, distributed control and optimization, and their applications to computational and systems neuroscience, connected and autonomous vehicles, and cyber-physical systems.

Professor Jiang is a Deputy Editor-in-Chief of the IEEE/CAA Journal of Automatica Sinica and of the Journal of Decision and Control and has served as Senior Editor for the IEEE Control Systems Letters (L-CSS) and Systems & Control Letters, Subject Editor, Associate Editor and/or Guest Editor for several journals including International Journal of Robust and Nonlinear Control, Mathematics of Control, Signals and Systems, IEEE Transactions on Automatic Control, European Journal of Control, and Science China: Information Sciences.

Leilei Cui is a third-year PhD student at the Department of Electrical and Computer Engineering, New York University, under the supervision of Professor Zhong-Ping Jiang. He received a B.S. Degree in Automation from Northwestern Polytechnical University, Xian, China, in 2016, and the M.S. degree in Control Science and Engineer from Shanghai Jiao Tong University, Shanghai, China, in 2019. His research interests are reinforcement learning, adaptive dynamic programming, control theory, and their applications to robotics and intelligent transportation.

Sayan Chakraborty is a first year PhD candidate at the Department of Electrical and Computer Engineering, New York University, under the supervision of Professor Zhong-Ping Jiang. He obtained a B.Tech. degree in Electrical Engineering from National Institute of Technology, Silchar, India in 2017, and an M.Tech. degree in Electrical Engineering with specialization in Systems and Control from Indian Institute of Technology Hyderabad, India in 2021. His research interests are data-driven control, adaptive dynamic programming, and their application to autonomous vehicles.