摘要
随着扫地机器人的快速发展,作为其核心技术的全覆盖路径规划技术也变得日益重要。目前已经提出的许多算法,如人工势场法、模板法、单元分解法等,都存在一些问题,如覆盖率低、重复率高、运行效率低等等。针对目前已有算法存在的问题,提出了一种基于回溯法的全覆盖路径规划算法。首先利用West-MoveFirst算法实现局部区域覆盖,然后为了解决扫地机器人局部区域覆盖过程中存在遗漏区域未覆盖的问题,建立了完善的回溯机制,并采用改进的A*算法规划出一条从死点到回溯点的光滑无障碍路径。通过与BA*算法进行仿真对比分析,表明了该算法具有更高的运行效率和更低的重叠率。
With the rapid development of sweeping robots,as the core technology,full coverage path planning technology has become increasingly important.Many algorithms have been proposed so far,such as the artificial potential field method,template method and unit decomposition method,however,they have problems such as low coverage,high repetition rate and low operating efficiency.In order to solve the problems existing in the proposed algorithms,we propose a full coverage path planning algorithm based on backtracking method.The algorithm firstly uses the west-move first algorithm to achieve local area coverage.Then,in order to solve the problem that the missing area is not covered in the process of local area coverage,we establish a perfect backtracking mechanism.And the improved A^*algorithm is used to plan a smooth unobstructed path from dead points to backtracking points.Simulation results show that compared with the BA^*algorithm,the proposed algorithm has higher operating efficiency and lower overlap rate.
作者
李楷
陈永府
金志勇
刘田
王振庭
郑迥之
LI Kai;CHEN Yong-fu;JIN Zhi-yong;LIU Tian;WANG Zhen-ting;ZHENG Jiong-zhi(School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China)
出处
《计算机工程与科学》
CSCD
北大核心
2019年第7期1227-1235,共9页
Computer Engineering & Science
作者简介
李楷(1995-),男,河南驻马店人,硕士生,研究方向为移动机器人即时定位和地图构建,移动机器人路径规划。E-mail:1368814300@qq.com;通信作者:陈永府(1972-),男,四川泸州人,博士生,讲师,研究方向为AGV、SLAM、数字水印和信息安全。E-mail:chenyf@mail.hust.edu.cn;金志勇(1994-),男,河北秦皇岛人,硕士生,研究方向为计算机漏洞攻击与利用,计算机内存管理及防护。E-mail:809291807@qq.com;刘田(1995-),男,陕西渭南人,硕士生,研究方向为SLAM和机器视觉。E-mail:2194167956@qq.com;王振庭(1994-),男,湖北黄冈人,硕士生,研究方向为多机器人路径规划与集群调度。E-mail:wanghappyu@qq.com.
