摘要
针对传统灰狼算法求解移动机器人路径规划问题收敛效率低且易陷入局部极值的缺陷,提出一种基于Tent混沌映射初始化种群的改进灰狼(TGWO)算法,并将其运用于解决移动机器人全局路径规划问题。基于Tent混沌映射初始化灰狼种群,以丰富种群多样性,提高收敛速度;提出指数型收敛因子改进策略,以更好地拟合灰狼实际搜索过程,并通过改进控制参数H以平衡算法的全局勘探与局部开发能力;融合动态权重因子和适应度比例系数,更新灰狼个体的位置信息,以提高灰狼个体自主搜索能力,避免算法陷入局部最优。为验证算法有效性,选用8个标准测试函数以及3组复杂度不同的栅格环境,先后开展了TGWO算法与传统GWO算法、3种典型改进灰狼算法的测试对比实验以及全局路径规划仿真对比实验。结果表明:TGWO算法在单峰、多峰函数上均有较好的收敛性、较高的寻优精度;仿真场景下,相较于传统GWO算法,TGWO算法所提的各个改进策略均能有效提升路径寻优性能;TGWO算法的平均路径长度、路径长度标准差、平均迭代次数、平均寻优耗时这4项指标均优于对比算法;TGWO算法路径寻优的优越性和鲁棒性得到了验证。
The traditional grey wolf algorithm achieves low convergence efficiency and tends to get struck in local extremum in solving path planning problems for mobile robots.As a result,this paper proposes an improved grey wolf algorithm(Tent-initialized grey wolf optimization,TGWO)based on the population initialization of Tent chaotic mapping,Firstly,the population initialization method based on Tent chaotic mapping is adopted to enrich the diversity of the population,which can improve the convergence speed.Secondly,the improvement strategy based on exponential convergence factor is proposed to better fit the search process of the grey wolf,and the global exploration and local exploitation capabilities of the algorithm are balanced by improving the control parameter H.Finally,the dynamic weight factor and the fitness scale coefficient are integrated to update the individual positions of the grey wolves,so as to improve the independent searching ability of the individual,and prevent the algorithm from getting trapped into local optimum.To verify the effectiveness of the proposed algorithm,experiments are carried out by comparing the TGWO,traditional GWO and three improved typical algorithms for global path planning simulation using eight standard test functions and three sets of grid environments with different complexities.The results are as follows:1)For both unimodal and multi-modal functions,the convergence performance and optimization accuracy of TGWO algorithm are better than others;2)Under the simulation scenes,compared with the traditional GWO,each improved strategy proposed for TGWO algorithm effectively improves the performance of path optimization;3)TGWO are superior to other algorithms in terms of the 4 indicators including the average path length,standard deviation of path lengths,average number of iterations and average time for optimization.All these verify the superiority and robustness of TGWO algorithm in path optimization.
作者
刘志强
何丽
袁亮
张恒
LIU Zhiqiang;HE Li;YUAN Liang;ZHANG Heng(School of Mechanical Engineering,Xinjiang University,Urumqi 830047,China;School of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China)
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2022年第10期49-60,共12页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(62063033,U1813220)。
关键词
移动机器人
路径规划
灰狼优化算法
Tent混沌映射
非线性控制参数
惯性权重系数
mobile robot
path planning
grey wolf optimization algorithm
Tent chaotic mapping
nonlinear control parameter
inertia weight coefficient
