Title: Reinforcement learning of dynamic collaborative driving Part II: lateral adaptive control
Authors: Luke Ng, Christopher M. Clark, Jan Paul Huissoon
Addresses: Department of Mechanical and Mechatronics Engineering, University of Waterloo, 200 University Ave W., Waterloo, Ontario, N2L 3G1, Canada. ' Computer Science Department, California Polytechnic State University, San Luis Obispo, CA, USA. ' Department of Mechanical and Mechatronics Engineering, University of Waterloo, 200 University Ave W., Waterloo, Ontario, N2L 3G1, Canada
Abstract: In dynamic collaborative driving, multiple vehicles coordinate their motion to optimise road usage using shared information. The basic prerequisites for a vehicle participating in dynamic collaborative driving are longitudinal and lateral control. This paper focuses on the lateral vehicle control on which higher-level manoeuvres such as entering or exiting a formation are based. Each vehicle involved is a composite nonlinear system powered by an internal combustion engine, equipped with automatic transmission, rolling on rubber tyres with hydraulic braking systems and steering system. A vehicle model is introduced which serves as the control system design platform. A lateral adaptive preview control system which uses Monte Carlo Reinforcement Learning (RL) is introduced. The results of the RL phase and the performance of the adaptive preview control system for a single automobile as well as the performance in a multi-vehicle platoon are presented.
Keywords: mobile robots; motion control; adaptive cruise control; collaborative driving; vehicle dynamics; vehicle simulation; machine learning; reinforcement learning; adaptive control; motion coordination; multi-vehicle platoon; lane-keeping; lateral adaptive control.
DOI: 10.1504/IJVICS.2008.022356
International Journal of Vehicle Information and Communication Systems, 2008 Vol.1 No.3/4, pp.229 - 248
Received: 28 Mar 2008
Accepted: 02 Apr 2008
Published online: 02 Jan 2009 *