基于改进蚁群算法的多农业机器人路径规划研究  

Research on Path Planning of Multi-agricultural Robots based on Improved Ant Colony Algorithm

作　　者：罗智杰 黄子涛 许嘉志 潘仲宇 曹亮 刘双印[1,2,3] Luo Zhijie;Huang Zitao;Xu Jiazhi;Pan Zhongyu;Cao Liang;Liu Shuangyin(College of Information Science and Technology,Zhongkai University of Agriculture and Engineering,Guangzhou 510225,China;Smart Agricultural Engineering Technology Research Center,Zhongkai University of Agriculture and Engineering,Guangzhou 510225,China;Guangzhou Key Laboratory of Information Technology for Traceability of Agricultural Products Quality and Safety,Guangzhou 510225,China)

机构地区：[1]仲恺农业工程学院信息科学与技术学院,广东广州510225 [2]仲恺农业工程学院智慧农业工程技术研究中心,广东广州510225 [3]广州市农产品质量安全溯源信息技术重点实验室,广东广州510225

出　　处：《现代农业装备》2021年第3期56-62,68,共8页Modern Agricultural Equipment

基　　金：国家自然科学基金(61871475);广东省科学计划项目(201905010006);广州市科学研究计划一般项目(201904010233,201905010006)。

摘　　要：针对蚁群算法在解决多机器人移动路径规划上存在的局限性,本文利用栅格法对环境进行空间建模,并在基本蚁群算法的基础上,对启发信息函数做出了改进,同时提出死锁问题的解决方案。改进方案既保证了算法的速度,又避免了算法的提前收敛。另外,针对未知运动状态和已知运动状态的障碍物,本文分别讨论了动态窗口搭配区域膨胀以及正碰、侧碰两种避碰策略,进一步加强了机器人应对多变的室内农业环境的能力。本文所研究改进的算法对于用在障碍物较多,路径规划较复杂的大型农场中有较好的效果,其有利于提高农业机器人在大面积农业场景的运作效率。The basic ant colony algorithm has some limitations in path planning.Based on the basic ant colony algorithm,this paper makes some improvements and solves the deadlock problem in the ant search process.In view of the possible existence of dynamic obstacles,this paper uses the method of rolling window to detect,and uses the regional expansion strategy to effectively avoid the occurrence of collision accidents.In addition,two common problems,forward collision and side collision,are solved by local path planning and parking waiting strategy.The improved algorithm studied in this paper has a good effect on large farms with many obstacles and complicated path.It is helpful to improve the operation efficiency of agricultural robot in large area agricultural scene.

关 键 词：蚁群算法 动态环境 多机器人 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]