Local path planning of intelligent vehicle based on improved artificial potential field

Abstract

The actual environment of vehicles will inevitably encounter moving obstacles such as pedestrians and vehicles, and the vehicles need to get back to the global path in time after avoiding moving obstacles. In order to avoid obstacles safely, artificial potential field is applied to local dynamic path planning. Aiming at solving the problems of traditional artificial potential field, the traditional artificial potential field is improved in this paper, which include discretizing the boundary of obstacles to ensure the safety of obstacle avoidance, adding random escape force to escape the local minimum and considering the speed and acceleration of obstacles to apply traditional artificial potential field to dynamic path planning. The design of obstacle avoidance for three most common collisions of front collision, rear collision and side collision is carried out. The improved artificial potential field is used to acquire the local path. The simulation results show that the proposed algorithm can obtain local dynamic paths with better safety and real-time performance. Combined with the global path, the path planning of intelligent vehicles is completed in this paper.