一种面向群组机器人搜索任务的改进粒子群算法  

An improved particle swarm optimization algorithm for group robots searching task

作　　者：康亮[1] 徐杨 尹丽华 KANG Liang;XU Yang;YIN Lihua(Engineering Training and Innovation Education Center,Shanghai Polytechnic University,Shanghai 201209,China;School of Computer and Information Engineering,Shanghai Polytechnic University,Shanghai 201209,China)

机构地区：[1]上海第二工业大学工程训练与创新教育中心,上海201209 [2]上海第二工业大学计算机与信息工程学院,上海201209

出　　处：《中国科技论文》2023年第11期1244-1249,共6页China Sciencepaper

基　　金：中国高校产学研创新基金资助项目(2021ZYA03008)。

摘　　要：针对群组机器人搜索时难以得到全部全局信息的问题,提出邻域粒子对粒子群优化(particle swarm optimization,PSO)算法进行改进。对于机器人组群中的粒子,设定了机器人的速度限制,根据任务的解决程度,可以弹性改变机器人粒子的前进速度。为实现在实际环境中使用PSO算法,将粒子的拓扑空间替换为搜索空间,使得群组机器人可以应用改进后的PSO算法完成既定的搜索任务。设计了一套含有8个指标的算法评价体系,通过100次的3种不同类型搜索任务求解,对比不同的算法和指标,实验结果证明了提出的PSO算法在群组机器人搜索任务中的适用性和有效性。Facing the challenge realted to the difficulty in the obtaining of all global information during group robot searching,neighborhood particles were proposed to improve the particle swarm optimization(PSO)algorithm in the present work.For the particles in the robot group,the speed limit of the robot was set.Based on the progress of the task,the robot’s forward speed could be flexibly changed.In order to use the PSO algorithm in the actual environment,the particle topological space was replaced by the search space,through which the group of robots could apply the improved PSO algorithm to complete the established searching tasks.Finally,a set of algorithm evaluation system with 8 indexes was designed.The applicability and effectiveness of the proposed PSO algorithm in group robots searching tasks is verified through 100 times of solving three different types of searching tasks and comparing different algorithms and different indicators.

关 键 词：群组机器人 粒子群优化(PSO)算法 搜索任务 邻域粒子 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]