基于改进多目标自适应遗传算法的机器人路径规划  

Robot path planning based on improved multi-objective adaptive genetic algorithm

作　　者：张铮[1] 柯子鹏 周嘉政 钱勤建 胡新宇[1] ZHANG Zheng;KE Zipeng;ZHOU Jiazheng;QIAN Qinjian;HU Xinyu(School of Mechanical Engineering,Hubei University of Technology,Wuhan 430068,China)

机构地区：[1]湖北工业大学机械工程学院,湖北武汉430068

出　　处：《西安理工大学学报》2023年第1期69-78,共10页Journal of Xi'an University of Technology

基　　金：国家自然科学基金资助项目(61976083)。

摘　　要：针对传统遗传算法收敛速度较慢、早熟,混合遗传算法复杂、耗时等不足,提出一种改进多目标自适应遗传算法。在初始化操作中,提出一种限制性均匀随机搜索算法结合中值插入算法初始化种群,通过均匀节点库随机生成节点,结合限制性步长控制节点搜索范围,并建立限制性步长与产生初始种群长度的先验模型。改进了自适应交叉变异操作,通过平衡阈值缩小其计算复杂度。利用自适应进化操作进化判断,同时缩短种群进化停滞过程,结合贪心算法防止种群出现倒退现象。最后,采用删除操作,平滑最优路径。通过与传统遗传算法(GA)、蚁群遗传算法(ACO-GA)、麻雀搜索算法(SSA)对比实验,仿真结果表明,改进的自适应遗传算法效率高,以更少次数收敛,具有更好的迭代稳定性,同时降低了机器人能耗。In view of the slow convergence speed and precociousness of traditional genetic algorithms,and the complex and time-consuming hybrid genetic algorithms,an Improved Multi-Objective Adaptive Genetic Algorithm is put forward for these deficiencies.In the population initialization operation,a restricted uniform random search algorithm combined with the median insertion algorithm is proposed to initialize the population,randomly generate nodes through a uniform node library and control the search range of nodes combined with the restrictive step size,with a priori model of restrictive step size established based on the initial path length.The adaptive cross-mutation operator is optimized by balancing the threshold to improve the whole efficiency by the algorithm.An adaptive evolution is put forward to operate evolutionary judgment while shortening the process of population evolution stagnation,combined with greedy algorithm to prevent population regression.Finally,a deletion operation is applied to optimize path.This paper improves the algorithm compared with the traditional Genetic Algorithm(GA),the Ant Colony Genetic Algorithm(ACO-GA)and the Sparrow Search Algorithm(SSA).The results display that the Improved Multi-Objective Adaptive Genetic Algorithm can converge with higher efficiency,and it has better performance.Its iterative stability is improved while reducing the energy consumption of robot.

关 键 词：平衡阈值 限制性均匀随机搜索 遗传算法 自适应进化 删除操作 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]