移动机器人导航路径的异构串行蚁群算法规划  

Heterogeneous Serial Ant Colony Based Planning Method of Mobile Robot Navigation Path

作　　者：蒋泽艳 郭林炀 廖军 周正平 JIANG Ze-yan;GUO Lin-yang;LIAO Jun;ZHOU Zheng-ping(College of Information Technology,Changchun Finance College,Jilin Changchun 130000,China;State Key Labora-tory of High Power Semconductor Laser,Changchun University of Science and Technology,Jilin Changchun 130022,China;College of Intelligent Manufacturing,Yangzhou Polytechnic Institute,Jiangsu Yangzhou 225000,China;Ji-angsu Shuguang Electro-Optics Co.,Ltd.,Jiangsu Yangzhou 225000,China)

机构地区：[1]长春金融高等专科学校信息技术学院,吉林长春130000 [2]长春理工大学高功率半导体激光国家重点实验室,吉林长春130022 [3]扬州工业职业技术学院智能制造学院,江苏扬州225000 [4]江苏曙光光电有限公司,江苏扬州225000

出　　处：《机械设计与制造》2024年第1期342-346,共5页Machinery Design & Manufacture

基　　金：江苏省高等学校自然科学研究项目资助(20KJB470032)。

摘　　要：为了减少机器人工作路径长度和拐点数量,提出了基于异构串行蚁群算法的路径规划方法。介绍并改进了栅格模型的投影、膨化和编码过程;以减少工作路径长度和拐点数量为目标,建立了机器人导航路径的优化模型。分析了标准蚁群算法原理,依据起点和目标点位置的先验知识,引入了信息素非均匀初始化方法;在概率公式中添加了角度启发因子;并在蚁群中加入具有全局通视能力的蚂蚁,提出了异构蚂蚁串行策略。在(10×10)和(20×20)规模栅格中,将异构串行蚁群算法用于机器人路径规划,与标准蚁群算法和文献[12]超强启发算法相比,异构串行蚁群算法规划路径长度最短,且拐点数量远小于另外两种算法。仿真结果表明异构串行蚁群算法在机器人路径规划中具有优越性。In order to reduce the length of robot working path and the number of inflection points,a path planning method based on heterogeneous serial ant colony algorithm is proposed.The projection,expansion and coding process of grid model are intro-duced and improved;In order to reduce the length of working path and the number of inflection points,the optimization model of robot navigation path is established.The principle of standard ant colony algorithm is analyzed.According to the prior knowledge of the position of starting point and target point,the pheromone non-uniform initialization method is introduced;The angle heu-ristic factor is added to the probability formula;The ants with global intervisibility ability are added to the ant colony,and the heterogeneous ant serial strategy is proposed.At(10×10)and(20×20)scale grid,the heterogeneous serial ant colony algorithm is used for robot path planning.Compared with the standard ant colony algorithm and the super heuristic algorithm in refer-ence[12],path length planned by heterogeneous serial ant colony algorithm is the shortest and the number of inflection points is much smaller than that of the other two algorithms.The simulation results show that advantages of heterogeneous serial ant colony algorithm in robot path planning are obvious.

关 键 词：移动机器人 路径规划 栅格模型 异构串行蚁群 角度启发因子 

分 类 号：TH16[机械工程—机械制造及自动化] TH162