Abstract: Aiming at the problem of obstacle avoidance path planning for intelligent vehicle, an obstacle avoidance path planning and design method based on improved artificial potential field method is proposed. Firstly, the vehicle speed information is introduced into the traditional repulsive potential field function, the lane center line and road boundary potential field are designed, and the path obstacle avoidance planning controller is designed; Then, aiming at the problem of intelligent vehicle path tracking control, a three-degree-of-freedom vehicle dynamics model is established, and the horizontal and longitudinal tracking controllers are designed by using linear time-varying model predictive control and back propagation-proportional-integral-differential (BP-PID) control algorithm, respectively, to track the lateral position and longitudinal velocity of the vehicle; Finally, based on the CarSim/Simulink co-simulation platform, the validity of the design method is verified in the double line shifting condition and the straight line condition. The simulation results show that the safe and reliable obstacle avoidance path can be obtained by using the improved artificial potential field method, and the designed horizontal and longitudinal tracking controller has good tracking performance.

Key words: Model predictive control; Artificial potential field method; Path tracking; Obstacle avoidance path planning; Neural network; CarSim/Simulink

摘要： 针对智能车辆避障路径规划问题，提出一种基于改进人工势场法的避障路径规划设计 方法，首先在传统斥力势场函数中引入车辆速度信息，并设计了车道中心线、道路边界势场， 以及路径避障规划控制器；然后针对智能车辆路径跟踪控制问题，建立了三自由度车辆动力 学模型，分别采用线性时变模型预测控制和反向传播-比例-积分-微分（BP-PID）控制算法设 计横、纵向跟踪控制器，对车辆的横向位置和纵向速度进行跟踪；最后基于 CarSim/Simulink 联合仿真平台，在双移线工况和直线工况下，验证设计方法的有效性。仿真结果表明，采用 改进的人工势场法可获得安全可靠的避障路径，所设计的横纵向跟踪控制器具有良好的跟踪 性能。

关键词: 模型预测控制；人工势场法；路径跟踪；避障路径规划；神经网络；CarSim/Simulink