Random Patrol Path Planning for Unmanned Surface Vehicles in Shallow Waters

Abstract

To solve the problem of surface patrol in small and medium-sized shallow waters, a multi-point path planning scheme for unmanned surface vehicle (USV) is proposed. The random patrol path planning method is designed as follows: 1) randomly select target points 2) combine genetic algorithm to solve Traveling Salesman Problem (TSP) 3) use Rapidly-exploring Random Trees Star algorithm (RRT*) and Dynamic Window Approach algorithm (DWA) to plan the path. Through the cooperation of global path planning and local path planning, under the premise of ensuring the safety of the USV, the random search of the water surface can be more efficiently carried out. Through simulation experiments, it is proved that the path planning algorithm designed in this paper can complete the patrol task of the water surface under the condition that the USV is safe in shallow waters.