Dynamic path planning for unmanned surface vehicle in complex offshore areas based on hybrid algorithm

Abstract

According to its planning scope, path planning for unmanned surface vehicle (USV) can be divided into global and local path planning. Many scholars have improved the classic algorithms, including grids method, visibility graph method, A* algorithm and artificial potential field method (APF), But the global planning algorithm still has outstanding problems such as long calculation time and large computational overhead in large task space, local planning algorithms usually ignore the global optimal constraints. Aiming at the problem of dynamic path planning of environmental monitoring USV under complicated offshore navigation conditions, based on the idea of bi-level planning, a hybrid algorithm which combines global and local path planning is proposed. This paper first proposes an improved Particle Swarm Optimization (PSO) for global path planning according to the given information about marine environment, and introduces Opposition-based Learning (OBL) and improves the inertia weight as well as search step size to effectively avoid the precocity of PSO. Then on the basis of global optimized path and sensor information, the improved Artificial Potential Field (APF) algorithm is adopted for local dynamic obstacle avoidance, so as to solve the local minimum problem. The results of simulation indicated that the improved PSO can effectively avoid the precocity of particles and enhance the optimization capability and stability of the PSO; the improved APF would not be restricted by local minimum point and achieve dynamic obstacle avoidance under the constraints of global optimization path. Therefore, the combination of these two algorithms can effectively solve the problems of path optimization and dynamic obstacle avoidance for environment monitoring USV when it is executing missions in complex offshore areas.