Vehicle Local Path Planning Based on Artificial Potential Field Method
Junwen Pan, Yi Jiang, Jin Mao · 2024
Local path planning is an important module of intelligent vehicles, and it is crucial to be able to complete obstacle avoidance safely and smoothly. Since the traditional artificial potential field method treats obstacles as mass points, after applying it to vehicle obstacle avoidance path planning, the planned path is close to the obstacles and the curvature of the road is large, which is not conducive to the tracking of the control module. To address this problem, the traditional artificial potential field method is improved by using quintic polynomial obstacle model instead of the mass point model, introducing road constraints, and combining with the particle swarm optimization to solve the optimal path. In order to obtain a smooth obstacle avoidance trajectory, the planned path is smoothed using Bezier curves. Finally, MATLAB is utilized for simulation, and by comparing with the traditional artificial potential field method, it is verified that the improved algorithm can better complete the obstacle avoidance path planning, and the road smoothness is improved.