A Hybrid Path-Planning Scheme for an Unmanned Surface Vehicle

Abstract

In this paper, a hybrid path-planning scheme that combines global A* algorithm with local dynamic window based collision avoidance is proposed for an unmanned surface vehicle (USV) in complex environment. Considering the way-points tracking in maritime applications, the A* algorithm with post-smoothed (A* PS) method is employed to reduce the number of way-points, and thereby contributing to plan a shortest path without constraining on grids. The local collision avoidance is realized by the Dynamic Window approach which takes the motion dynamics of the USV into account. Furthermore, a virtual safety zone pertaining to the shape of the obstacle is established to ensure reliable navigation at high speed. Simulation studies demonstrate that the proposed global-local hybrid path-planning scheme achieves remarkable performance and superiority in path planning with obstacle avoidance.