Alexandre Bayen (Berkeley) Deep Reinforcement Learning for Vehicle Control

7/30/2019
Abstract:
For large scale inference and control of traffic flow, we consider fluid flow models over networks and discuss algorithms relying on convex optimization for fusion of static and mobile (Lagrangian) traffic information. We characterize the stability of such flows under selfish information patterns (each flow attempting to optimize its latency) and show convergence to Nash equilibria leading to control strategies optimizing network efficiency.

We discuss how self-driving vehicles will change traffic flow patterns at local scale and describe approaches based on deep-RL applied to mixed-autonomy mobility. The talk explores the gradual and complex integration of automated vehicles into the existing traffic system. We examine the potential impact of a small fraction of automated vehicles on low-level traffic flow dynamics using novel techniques in model-free deep reinforcement learning in which the automated vehicles act as mobile controllers to traffic flow.


Bio:
Alexandre Bayen is the Liao-Cho Professor of Engineering at UC Berkeley. He is a Professor of Electrical Engineering and Computer Science and Civil and Environmental Engineering. He is currently the Director of the Institute of Transportation Studies (ITS). He is also a Faculty Scientist in Mechanical Engineering at the Lawrence Berkeley National Laboratory (LBNL). He received the Engineering Degree in applied mathematics from the Ecole Polytechnique, France, in 1998, the M.S. and Ph.D. in aeronautics and astronautics from Stanford University in 1998 and 1999 respectively. He was a Visiting Researcher at NASA Ames Research Center from 2000 to 2003. From January to December 2004, he worked as the Research Director of the Autonomous Navigation Laboratory at the Laboratoire de Recherches Balistiques et Aerodynamiques. He has been on the faculty at UC Berkeley since 2005 and has authored two books and over 200 articles.