重庆理工大学学报(自然科学) ›› 2024, Vol. 38 ›› Issue (1): 355-367.

• “智能机器人感知、规划及应用技术”专栏 • 上一篇    下一篇

改进海洋捕食者算法的机器人路径规划研究

黄训爱,杨光永,樊康生,徐天奇   

  1. 云南民族大学电气信息工程学院,云南省无人自主系统重点实验室
  • 出版日期:2024-02-07 发布日期:2024-02-07
  • 作者简介:黄训爱,男,硕士研究生,主要从事路径规划研究,Email:2443882889@qq.com;通信作者杨光永,男,博士,副教授,主要从事算法设计、信号处理、智能控制研究。本文引用格式:黄训爱,杨光永,樊康生,等.改进海洋捕食者算法的机

Robot path planning research with improved marine predator algorithm

  • Online:2024-02-07 Published:2024-02-07

摘要: 为解决海洋捕食者算法(MPA)收敛速度慢、收敛精度低、易陷入局部最优的问题,提出一种多策略改进海洋捕食者算法(IMPA)。引入Logistic混沌映射初始化种群,增加捕食者种群多样性;基于当前迭代次数t的自适应移动步长动态调整策略,增强算法逃离局部最优的能力;在IMPA迭代后期,加入中垂线算法(MA),基于中垂线策略的游离粒子位置更新方法,能够加快更新捕食者的位置,增强算法的寻优速度和寻优精度,避免算法陷入局部最优。最后通过改变IMPA阶段转换寻优过程,进一步平衡搜索过程,加强全局与局部适应性。选用6个基准测试函数对算法性能进行测试,测试结果显示:IMPA收敛速度更快,收敛精度更高;最后将改进算法应用于移动机器人路径规划,仿真结果表明:该算法规划的路径长度更短,搜索效率更高。

关键词: 混沌映射, 中垂线算法, 移动步长, 游离粒子, 路径规划

Abstract: This paper proposes a multi-strategy Improved Marine Predator Algorithm (IMPA) to address the Marine Predator Algorithm’s (MPA) problems of slow convergence, low convergence accuracy, and tendency to fall into local optimum. The proposed algorithm introduces a logistic chaotic mapping to initialize the population and increase the diversity of the predator population; an adaptive moving step dynamic adjustment strategy based on the current iteration number t enhances the ability of the algorithm to escape from local optimum; a mid-pipeline algorithm (MA) is added in the late iteration of IMPA, and a free particle position update method based on the mid-pipeline strategy accelerates the position update of the predator and enhances the algorithm’s accuracy and optimization search speed, avoiding falling into local optimum. Finally, the search process is further balanced and the global and local adaptability is enhanced by changing the IMPA stage to transform the search process. Six benchmark test functions are selected to test the performance of the algorithm. Our test results show IMPA converges faster and achieves a higher convergence accuracy. Finally, the improved algorithm is applied in mobile robot path planning, and the simulation results show the algorithm plans a shorter path and achieves a higher search efficiency.

中图分类号: 

  • TP301.6