改进海洋捕食者算法的机器人路径规划  

Robot path planning based on improved MPA

作　　者：戚得众[1,2] 田晨 袁丽峰 吴云志 郑拓 QI Dezhong;TIAN Chen;YUAN Lifeng;WU Yunzhi;ZHENG Tuo(School of Mechanical Engineering,Hubei University of Technology,Wuhan 430068,China;School of Mechanical and Electronic Engineering,Shandong Agricultural University,Taian 271018,China)

机构地区：[1]湖北工业大学农机工程研究设计院,湖北武汉430068 [2]山东农业大学机械与电子工程学院,山东泰安271018

出　　处：《现代电子技术》2024年第24期153-159,共7页Modern Electronics Technique

基　　金：国家重点研发计划项目(2018YFD0700604)。

摘　　要：为克服海洋捕食者算法存在的初始种群分布不均、收敛速度慢且易陷入局部最优等问题,提出一种改进的海洋捕食者算法(GMPA)。首先,在初始化种群时采用Sobol序列和对立学习相结合的策略,得到更加均匀随机的初始解;再引入惯性权重系数和柯西变异来更新种群,提高算法跳出局部最优的能力;最后,针对更新后的种群,结合随机性学习策略来增加迭代过程中种群的多样性。为验证所提算法的有效性,选用7个标准测试函数对其性能进行评估;并选用3组复杂度不同的栅格地图,对改进后的算法与原始算法开展路径规划对比实验。实验结果表明:改进后的海洋捕食者算法在机器人路径规划问题中表现出良好的性能,显著提高了收敛速度并增强了优化能力。In allusion to the shortcomings of the marine predator algorithm(MPA),such as uneven distribution of the initial population,slow convergence speed and easy to fall into local optimum,an improved MPA is proposed.A combination of Sobol sequence and contrastive learning strategy is used in initializing the population to obtain a more uniform and random initial solution.The inertia weight coefficient and Cauchy's variance are introduced to update the population,so as to improve the the algorithm's ability to escape from local optima.The stochastic learning strategy is combined with the updated population to increase the diversity of the population in the iterative process.In order to verify the effectiveness of the algorithm,7 standard test functions are selected to evaluate the performance of the improved algorithm.The comparative experiments on path planning between the improved algorithm and the original algorithm are conducted by selecting 3 groups of raster maps with different complexities.The experimental results show that:the improved MPA has shown good performance in robot path planning problems,significantly improving convergence speed and enhancing optimization capabilities.

关 键 词：路径规划 海洋捕食者算法 Sobol序列 对立学习策略 种群分布 随机性学习策略 路径寻优 

分 类 号：TN820.4-34[电子电信—信息与通信工程] TP242[自动化与计算机技术—检测技术与自动化装置]