Improved artificial potential field algorithm for intelligent vehicle path planning

Abstract

Artificial Potential Field (APF) algorithm is one of the algorithms in the field of intelligent vehicle path planning. It can plan a safe path without collision for intelligent vehicle during driving from the start station to target station. However, the original APF algorithm also has some limitations. By changing the direction of the repulsive force and setting the initial conditions, this paper solves the inherent shortcoming of original APF algorithm, such as the target position is unreachable and easily fall into the local optimum during path planning. And due to the actual obstacle is not a particle, the obstacle safety distance is also proposed in this paper. Simulation results in MATLAB software show that the improved APF algorithm is effective. Then the improved APF algorithm is applied to the field of the intelligent vehicles path planning. The path following simulations in CARSIM software verify the followability of the target path which is planned by the improved APF algorithm. Therefore, the improved APF algorithm is effective and feasible for intelligent vehicle path planning.