Tracking Control for Autonomous Four-Wheel Independently Driven Vehicle Based on Deep Reinforcement Learning

This article proposes a deep reinforcement learning-based trajectory tracking control method for autonomous four-wheel independently driven vehicles. The method adopts a hierarchical structure. In the upper layer, a deep deterministic policy gradient (DDPG) algorithm is applied to enhance the vehicle's lateral stability during trajectory tracking to determine the optimal front wheels' steering angle and external yaw moment. In the lower layer, the external yaw moment input is calculated by the quadratic programming (QP) algorithm to obtain each in-wheel motor's(IWMs) torque output. The simulation results show that the algorithm can simultaneously provide tracking performances and ensures the lateral stability of autonomous vehicles.