摘要
采用蚁群算法求解移动机器人路径规划时,会出现收敛速度慢、搜索精度不高等问题。针对以上不足,首先,在传统蚁群算法的基础上对初始信息素采取非均匀式分配,避免蚂蚁进行无用的搜索行为,提高求解速度;其次,引入A*算法的启发搜索来改进蚁群算法的启发函数,加快搜索速度;然后,改进转移概率解决了死锁现象;最后,采用蚂蚁回退策略处理U型陷阱。MATLAB仿真结果表明,改进后的蚁群算法迭代次数减少了34%,搜索时间降低了60%,规划出的路径缩短了7%。
When the ant colony algorithm is used to solve the path planning of mobile robots,problems such as slow convergence speed and low search accuracy will occur.In view of the above shortcomings,firstly,on the basis of the traditional ant colony algorithm,the initial pheromone is distributed unevenly to avoid the useless search behavior of ants and improve the solution speed;secondly,the heuristic search of A*algorithm is introduced to improve the heuristic function of ant colony algorithm and speed up the search speed;then,the deadlock phenomenon is solved by improving the transition probability;finally,the ant fallback strategy is used to deal with U-shaped traps.The MATLAB simulation results show that the iteration times of the improved ant colony algorithm are reduced by 34%,the search time is reduced by 60%,and the planned path is shortened by 7%.
作者
周敬东
杨磊
高伟周
汪宇
ZHOU Jingdong;YANG Lei;GAO Weizhou;WANG Yu(School of mechanical Engin.,Hubei Univ.of Tech.,Wuhan 430068,China;Hubei Agricultural Machinery Engin.Research and Design Institute,Wuhan 430068,China)
出处
《湖北工业大学学报》
2021年第5期19-22,共4页
Journal of Hubei University of Technology
基金
“十三五”国家重点研发计划智能农机装备专项(2017YFD0700905)。
