A Path Planning and Tracking Control for Autonomous Vehicle With Obstacle Avoidance

This paper presents a path planning and tracking framework to implement obstacle avoidance for autonomous car. The safe driving area model, which is made up of a serious of nodes, designed by using the longitudinal and lateral motion relationship of the vehicle, is proposed to describe the location of automobile with sideslip constraint in driving and position of the obstacles on the road, and Q-learning algorithm is used to learn the optimal strategy on each node and the optimal path is obtained under the predefined rules. In order to execute precise path tracking control, linear output regulation method is introduced to design the tracking controller based on both the kinematic and dynamic vehicle models. CarSim simulations are conducted with different scenarios and the effectiveness of the proposed framework is demonstrated.