基于改进A^(*)算法和系统短期状态预测的仓储AGV路径规划方法  

Warehouse AGV path planning method based on improved A^(*)algorithm and system short-term state prediction

作　　者：王云峰 曹小华[1] 郭兴[1] WANG Yunfeng;CAO Xiaohua;GUO Xing(School of Logistics Engineering,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]武汉理工大学物流工程学院,湖北武汉430063

出　　处：《计算机集成制造系统》2023年第11期3897-3908,共12页Computer Integrated Manufacturing Systems

基　　金：国家自然科学基金资助项目(61503291);国家重点研发计划资助项目(2018YFC1407405)。

摘　　要：针对目前仓储自动导引车(AGV)路径规划中存在的转弯次数多、冲突节点多的问题,首先针对单AGV提出一种基于障碍预测的改进A^(*)算法,引入“AGV方向感”概念,通过判断和比较目标位置和AGV所在位置,同时结合障碍矩阵,在传统A^(*)算法中的启发函数中加入“方向函数”,使AGV可以提前预知障碍而另寻最佳路径;其次在单AGV路径规划基础上考虑减少路径中冲突节点数量,提出一种基于系统短期状态预测的多AGV路径规划方法;最后搭建仿真实验平台验证改进A^(*)算法和路径规划方法的有效性。实验结果表明,改进的A^(*)算法和基于短期状态预测多AGV路径规划方法可以有效减少转弯次数和路径中冲突节点数量,降低任务总完成时间。To solve the problems of multiple turns and conflicting nodes in the logistics and warehousing Automated Guided Vehicles(AGV)path planning,an improved A^(*)algorithm based on obstacle prediction was proposed for a single AGV,and the concept of"AGV sense of direction"was introduced to judge and compare the location of the target and the AGV.An obstacle matrix and a"direction function"were added to the heuristic function in the traditional A^(*)algorithm to make AGV predict obstacles in advance and find another optimal path.On the basis of the single AGV path planning,it was considered to reduce the number of conflicting nodes in the path,and a multi-AGV path planning method based on system short-term state prediction was proposed.A simulation experiment platform was built to verify the effectiveness of the improved A^(*)algorithm and path planning method.The experimental results showed that the improved A^(*)algorithm and the multi-AGV path planning method based on short-term state prediction could effectively reduce the number of turns and conflicting nodes in the path,and reduce total task completion time.

关 键 词：仓储自动导引车 路径规划 改进A^(*)算法 短期状态预测 冲突节点 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]