Research on Path Planning Method of Unmanned Boat Based on Improved Artificial Potential Field Method

Abstract

Aiming at the problem that the path planning algorithm of Unmanned Surface Vehicle (USV) is difficult to take into account the unequal dynamic and static environment, this paper proposes an improved Artificial Potential Field (APF). Aiming at the defects of unreachable target and local minimum in traditional APF, starting from the obstacle detection model and the USV motion characteristic constraints, the potential field model is modified by introducing a Gaussian function about the target distance, so that the USV can reach the nearby obstacle. The target point, and a repulsive deflection model is proposed to predict the magnitude of the resultant force on the USV, and gradually change the direction of the repulsive force on the USV to escape the “local trap”. The A* algorithm is proposed to preplan the map environment and extract the key turning points, which can be used as the sub-target points of the improved APF algorithm after the corresponding offset, so as to solve the problem of path optimization. The simulation results show that after improving the APF, the path length is reduced by 8.49% on average; the planned path is smoother, and the smoothness is reduced by 60.62% on average; the dangerous nodes of the planned path are reduced, and the safety degree is reduced by 55.15% on average. Improving APF can solve the defects of traditional APF, which is of great significance to improve the work efficiency and work safety of USV.