水面无人艇的多方向路径规划方法  

Multi-direction Path Planning Method of Surface Unmanned Vehicle

作　　者：杜贻群 黄静[2] 张华军[2] DU Yi-qun;HUANG Jing;ZHANG Hua-jun(Unit 92942 of PLA,Beijing 100161;Wuhan University of Technology,Wuhan 430070,China)

机构地区：[1]中国人民解放军92942部队,北京100161 [2]武汉理工大学,湖北武汉430070

出　　处：《指挥控制与仿真》2021年第4期7-12,共6页Command Control & Simulation

摘　　要：针对水面无人艇直航最短路径优化问题,采用栅格法进行二维环境建模以分离障碍物与可行域,在此基础上,对传统Dijkstra算法的四方向搜索进行改进,设计了一种多方向路径搜索算法,不仅使算法搜索速度得到大幅提高,还能满足无人航行器在实际航行时具备任意方向航向的要求。结合实际海洋环境对所提出的多方向搜索算法性能进行对比与分析,测试结果表明,多方向路径搜索算法比基本的四方向搜索算法获得的路径更短、转折角更少、转向更平滑,符合实际应用需求。Aiming at the shortest path optimization problem of surface unmanned vehicle direct navigation,grid method is used to model the two-dimensional environment to separate obstacles and feasible region.On this basis,it improves the four direction search of traditional Dijkstra algorithm,and designs a multi-direction path search algorithm,which not only greatly improves the search speed of the algorithm,but also meets the requirements of unmanned aerial vehicle in actual navigation direction and heading requirements.Combined with the actual marine environment,the performance of the proposed multi-directional search algorithm is compared and analyzed.The test results show that the multi-directional path search algorithm obtains shorter path,less turning angle and smoother steering than the basic four direction search algorithm,which meets the practical application requirements.

关 键 词：水面无人艇 栅格法 DIJKSTRA算法 路径规划 

分 类 号：U664.82[交通运输工程—船舶及航道工程] U666[交通运输工程—船舶与海洋工程]