混合模拟退火与粒子群优化算法的无人艇路径规划  

A Hybrid Optimization Algorithm of Simulated Annealing and Particle Swarm for Unmanned Surface Vessel Path Planning

作　　者：郑佳春[1] 吴建华[2] 马勇[2] 龙延[1] 

机构地区：[1]集美大学,福建厦门361021 [2]武汉理工大学,湖北武汉430063

出　　处：《中国海洋大学学报（自然科学版）》2016年第9期116-122,共7页Periodical of Ocean University of China

基　　金：国家自然科学基金项目(51309186)"复杂海况下的海事无人艇路径规划研究";福建省科技计划重点项目(2014H0036);福建省自然科学基金项目(2013J01203)资助~~

摘　　要：针对水面无人艇路径规划中单独使用模拟退火算法存在的不足,以提高无人艇在执行任务时的安全性为研究目的,解决传统无人艇在航行中未考虑来往船只而导致的碰撞危险,提出了模拟退火算法与粒子群算法相结合的混合路径规划算法,利用该算法的高收敛性和容易跳出局部最优等特点,结合避碰规则,实现海事无人艇在同一静态环境下对遇、交叉和追越三种会遇局面的最优路径规划。仿真结果表明:本文提出的算法可以实现无人艇在复杂水域条件下快速路径规划,使与他船的自动避碰行为成为可能,给出的路径规划具有可行、有效,能够为无人艇安全的航行、顺利的执行任务提供保障。With the exploration and development of marine resources,Unmanned Surface Vessel（USV）has become a hot research topic.Path planning is one of the key technologies of USV autonomous navigation,and its core technology is Planning Algorithm Path（PPA）.This paper analyses and studies several common path planning algorithms,discusses their advantages and disadvantages.The Particle Swarm Optimization（PSO）algorithm and Simulated Annealing（SA）algorithm are our main research object,and it is insufficient by using the PSO algorithm or SA algorithm alone for the existence of unmanned surface vehicle path planning.In order to improve the safety of USV sailing,to solve the traditional unmanned boat without considering the collision danger caused by the coming and going vessels during the voyage,the hybrid path planning algorithm（Hybrid of Simulated Annealing and Particle Swarm Optimization Algorithm,HSAPSO）combining SA algorithm and PSO algorithm is proposed.The SAPSO using the algorithm characteristics of convergence and easy to jump out of local optimum,combined with the collision regulations,it can be easy implementation the optimal path planning of USV in the three encounter situations of head on,crossing and overtaking,especial for Marine Unmanned Surface Vessel（MUSV）.The simulation results show that the SAPSO algorithm proposed in this paper enable the MUSV to do automatic collision avoidance behavior in complex water conditions,and the given path planning is feasible and effective.Which can provide safeguard for the MUSV＇s safe navigation and successful implementation of the task.

关 键 词：混合模拟退火与粒子群算法 无人艇 最优路径规划 自动避碰 

分 类 号：S948[农业科学—水产养殖] S943[农业科学—水产科学]