基于遗传-粒子群优化算法的USV路径规划方法

宫月红，张少君，王明雨，孟雄飞

山东交通学院 航运学院，山东 威海  264209

摘要：为解决传统粒子群优化算法（particle swarm optimization algorithm, PSO）应用于无人水面舰艇（unmanned surface vessel，USV）路径规划时存在的早熟收敛问题，提出一种结合遗传思想的PSO，在传统的PSO中引入遗传算法（genetic algorithm，GA）中的交叉、变异操作，避免算法进入局部最优解，对惯性权重进行自适应调整，加速算法收敛。采用MATLAB软件对USV巡检水域环境进行建模，应用改进的PSO进行路径规划。仿真结果表明：相对于传统的PSO和GA，该算法有效减少路径交叉点，大幅缩短路径总长和算法收敛时间。

关键词：USV；路径规划；PSO；GA；交叉变异

USV path planning method based on GA-PSO

GONG Yuehong,  ZHANG Shaojun,  WANG Mingyu,  MENG Xiongfei

Shipping College, Shandong Jiaotong University, Weihai 264209, China

Abstract:In order to solve the problem of precocity convergence when traditional particle swarm optimization algorithm(PSO) is applied to unmanned surface vessel (USV) path planning, a PSO algorithm based on genetic algoritdm(GA) is proposed. In this algorithm, the crossover and mutation operations of GA are introduced into the traditional PSO to avoid the algorithm entering local optimal solution. The convergence of the algorithm is accelerated by adaptively adjusting of the inertia weight. MATLAB simulation tool is used to model the environment of USV patrol area, and the improved PSO is used to carry out path planning. Simulation results show that, compared with traditional PSO and GA, the proposed algorithm can effectively reduce the path intersection points, thus greatly shorten the overall path length and the convergence time of the algorithm.

Keywords:USV; route planning; PSO; GA; crossover and mutation