摘要
鼠群优化(RSO)算法求解复杂环境下移动机器人路径规划问题时易出现早熟现象导致求解精度不足,针对此问题,提出一种多策略改进鼠群优化(MRSO)算法。首先,提出一种旋转小孔成像反向学习搜索策略,将其嵌入算法攻击猎物过程中对种群进行反向学习,提高算法全局搜索能力;其次,引入Iterative混沌RPOBL反向学习策略保证了算法的初始种群多样性,提高了算法初始寻优效率与收敛精度;最后,在算法追逐猎物过程中,采用“双平滑”和“双碗式”非线性自适应因子动态平衡了算法的全局搜索与局部探索,增强了算法局部和全局寻优能力。结果表明,在不同地图环境中,MRSO算法的路径寻优结果优于RSO、TSO和GWO算法,MRSO算法可快速和高效地解决复杂环境中移动机器人路径规划问题。
The rat swarm optimization(RSO)algorithm is prone to premature phenomenon when solving the path planning problem of mobile robots in complex environments,resulting in insufficient solution accuracy.In view of this,a multi-strategy rat swarm optimization(MRSO)algorithm is proposed.Firstly,a rotating pin-hole imaging opposition-based learning(RPOBL)is proposed,which is embedded in the algorithm to perform reverse learning on the population in the process of attacking the prey,so as to enhance the global search ability of the algorithm.Secondly,the Iterative chaos map with RPOBL mechanism is introduced to enrich the initial population of the algorithm to ensure the efficiency and accuracy at initial iteration phase.Finally,in the process of chasing the prey,the double-smooth and double-bowl nonlinear adaptive factors are used to dynamically balance the global search and local search of the algorithm to enhance the local and global optimization capabilities of the algorithm.The results show that MRSO algorithm is superior to RSO、TSO and GWO algorithm in path optimization in different map environments,and MRSO algorithm can quickly and efficiently solve the path planning problem of mobile robot in complex environment.
作者
解瑞云
海本斋
XIE Rui-yun;HAI Ben-zhai(School of Cable Engineering,Henan Institute of Technology,Xinxiang 453000,China;Faculty of Education,Henan Normal University,Xinxiang 453000,China)
出处
《组合机床与自动化加工技术》
北大核心
2022年第10期50-54,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
河南省高等学校重点科研项目(22B520006)
新乡市软科学研究计划项目(RKX2019013,RKX2021018)
河南省教师教育课程改革重点项目(2019-JSJYZD-013)
河南省本科高等学校智慧教学专项研究项目
河南工学院教学改革项目(2021-YB026)。
