基于粒子群联盟的机器人多目标搜索路径控制  

Path Control of Robot Multi-objective Search Based on Stand Subgroup Coalition

作　　者：吴琼 张红强[2] WU Qiong;ZHANG Hong-qiang(School of Electrical and Electronic Engineering,Guangdong University of Technology,Zhaoqing Guangdong 526100,China;School of Information and Electrical Engineering,Hunan University of Science and Technology,Xiangtan Hunan 411100,China)

机构地区：[1]广东理工学院电气与电子工程学院,广东肇庆526100 [2]湖南科技大学信息与电气工程学院,湖南湘潭411100

出　　处：《计算机仿真》2023年第8期426-429,529,共5页Computer Simulation

基　　金：湖南省自然科学基金项目(2021JJ30280);广东省2021省级一流本科课程(线上线下混合式一流课程)—自动控制原理(YLKC202102);广东理工学院2022创新强校工程科研项目(2022GKJZD005);广东理工学院2022质量工程项目(JXGG202201)。

摘　　要：在复杂环境下,机器人受未知障碍物影响,无法准确融合全部待搜索目标位置信息,在避障的同时无法精准实现任务分工。为解决上述问题,提出一种基于粒子群联盟的机器人多目标搜索路径控制方法。群机器人在并行化同时搜索多个目标时,根据机器人任务分工建立粒子群联盟,并对其确定目标协同搜索。通过改进自组织神经映射网络,在各个竞争神经元上组建锁定机制;同时采用自适应动态窗口,更新竞争神经元的权重,在障碍物信息未知的情况下实现避障,优化机器人的运动轨迹,从而实现多目标搜索路径协同控制。仿真结果表明,所提方法能够准确搜索多个目标,有效躲避障碍物,获取满意的协同控制结果。In a complex environment,the robot is affected by unknown obstacles,so it is unable to accurately fuse all the target location information to be searched,and it is unable to accurately achieve task division while avoiding obstacles.In order to solve this problem,a multi-objective search path control method for robots based on particle swarm alliance is proposed.When swarm robots search for multiple targets in parallel,they establish a sub group alliance according to the division of robot tasks,and determine the target cooperative search.By improving the self-organizing neural mapping network,a locking mechanism is established on each competing neuron;At the same time,the adaptive dynamic window is used to update the weight of the competitive neuron,realize obstacle avoidance under the condition of unknown obstacle information,and optimize the robot's motion trajectory,so as to achieve the multi-objective search path cooperative control.The simulation results show that the proposed method can accurately search multiple targets,effectively avoid obstacles,and obtain satisfactory cooperative control results.

关 键 词：复杂环境 群机器人 多目标搜索 路径 协同控制 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]