Obstacle-Avoiding Path Planning and Tracking for Intelligent Vehicles Based on Kinematic Model

With the rapid development of intelligent driving technology, people have higher requirements for the safety of vehicle driving. The framework of this paper is divided into two layers. The upper layer is responsible for the path planning of the driving trajectory, and the lower layer is the trajectory tracking control. In the path planning layer, an obstacle model based on the potential energy function is proposed, while considering the sensing range of the vehicle, so that the vehicle can make an obstacle avoidance response in advance. The search efficiency is improved by combining PSO to segment the trajectory. The differential flatness of the intelligent vehicles is used to smooth the whole trajectory and segment boundaries, making the driving more comfortable. The trajectory tracking layer utilizes the kinematic model and ensures that the inputs are within the control boundaries. Simulation and real experiment based on the Quanser self-driving car (QCar) have been accomplished to prove the effectiveness of the proposed algorithm.