基于IMA的AGV群组路径规划仿真  

AGV Group Path Planning Simulation Based on Improved Memetics Algorithm

作　　者：王海霞 甘卫华 尤凤翔 WANG Hai-xia;GAN Wei-hua;YOU Feng-xiang(Applied Technology College of Soochow University,Suzhou Jiangsu 215325,China;School of Transportation and Logistics,East China Jiaotong University,Nanchang Jiangxi 330013,China)

机构地区：[1]苏州大学应用技术学院,江苏苏州215325 [2]华东交通大学交通运输与物流学院,江西南昌330013

出　　处：《计算机仿真》2023年第12期200-208,542,共10页Computer Simulation

基　　金：国家自科基金(72061013)。

摘　　要：在大型仓储AGV群组作业任务场景中,路径规划带来的效率和安全是亟待解决的难题。针对传统文化基因算法(MA)容易陷入局部最优、耗时较长、路径不平滑等问题,在满足多约束条件下提出了一种改进的文化基因算法(IMA)。算法采用改进K聚类算法对环境栅格地图进行分区,缩小地图规模降低算法更新时间成本;根据适应度函数值采用自适应技术调整交叉和变异算子概率,增加种群多样性避免全局搜索陷入局部最优;通过概率法进入二次局部搜索,局部采用A*和蚁群混合算法改善规划路径的平滑性。最终提高了路径规划的效率和安全性。经验证IMA算法与传统MA算法相比提高了规划效率和避障性能,任务总时间平均节约4.6%,路径总长度节约4.3%,AGV能耗降低了27.1%,算法优化效果明显,适用于大型仓储AGV群组作业场景下的路径规划。In AGV group operation scenario of large warehouse,the efficiency and safety brought by path planning are urgent problems to be solved.Aiming at the problems of traditional cultural genetic algorithm(MA),such as easy to fall into local optimization,time-consuming and uneven path,an improved cultural genetic algorithm(IMA)was proposed under multiple constraints.The improved K clustering algorithm was used to partition the environmental ras⁃ter map and reduce the map size to reduce the algorithm update time cost.According to the fitness function,the prob⁃ability of crossover and mutation operator was adjusted by adaptive techniques to increase population diversity and a⁃void global search falling into local optimum.The probability method was used to enter the quadratic local search,and the A∗and ant colony hybrid algorithm was used to improve the smoothness of the planned path.Finally,the ef⁃ficiency and security of path planning were improved.Compared with the traditional MA algorithm,the IMA algorithm improves the planning efficiency and obstacle avoidance performance.The total task time is saved by 4.6%on aver⁃age,the total path length is saved by 4.3%,and the ENERGY consumption of AGV is reduced by 27.1%.The opti⁃mization effect of the algorithm is obvious,and it is suitable for path planning in AGV group operation scenarios of large warehouses.

关 键 词：路径规划 改进文化基因算法 外部框架算法改进 二次局部混合算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]