Reinforcement learning of dynamic collaborative driving Part I: longitudinal adaptive control

Abstract

Dynamic collaborative driving involves the motion coordination of multiple vehicles using shared information from vehicles instrumented to perceive their surroundings in order to improve road usage and safety. A basic requirement of any vehicle participating in dynamic collaborative driving is longitudinal control. Without this capability, higher-level coordination is not possible. Each vehicle involved is a composite non-linear system powered by an internal combustion engine, equipped with automatic transmission, rolling on rubber tyres with a hydraulic braking system. This paper focuses on the problem of longitudinal motion control. A longitudinal vehicle model is introduced which serves as the control system design platform. A longitudinal adaptive control system that uses Monte Carlo Reinforcement Learning (RL) is introduced. The results of the RL phase and the performance of the adaptive control system for a single automobile, as well as the performance in a multi-vehicle platoon, are presented.