基于双向A^(*)算法的自动导引车全局路径规划研究  

Research on global path planning for automated guided vehicles based on bidirectional A^(*) algorithm

作　　者：苏木雄 叶树林[1] SU Muxiong;YE Shulin(School of Mechatronic Engineering and Automation,Foshan University,Foshan 528225,China)

机构地区：[1]佛山科学技术学院机电工程与自动化学院,广东佛山528225

出　　处：《佛山科学技术学院学报（自然科学版）》2024年第3期20-26,共7页Journal of Foshan University(Natural Science Edition)

摘　　要：针对较大环境下传统A^(*)算法全局路径规划,存在搜索节点多、算法运算时间长、路径曲折等问题,提出了一种改进的双向A^(*)算法,该方法引入人工势场法的引力思想,使得搜索更具方向性;加入转弯代价,让路径更平滑;采用障碍因子,倾向选择宽敞的路径扩展。以路径长度、算法运算时间、搜索节点数、转弯次数的平均值作为评判指标,对比传统A^(*)算法、双向A^(*)算法及未采用障碍因子的改进算法,进行了8次仿真计算。仿真结果表明,改进A^(*)算法比双向A^(*)算法搜索节点少75.9%,转弯次数少53.6%,算法运算时间快84.1%。In traditional A^(*)algorithm-based global path planning for larger environments,issues such as an excessive number of search nodes,lengthy algorithm execution times,and convoluted paths arise.To address these challenges,an improved bidirectional A^(*)algorithm is introduced in this study.This approach incorporates the concept of artificial potential fields to provide a sense of direction in the search process.Additionally,a turning cost is integrated to promote smoother paths,while obstacle factors are introduced to favor the selection of broader path expansions.Finally,eight simulation runs are conducted to compare the proposed improved A^(*)algorithm with traditional A^(*)and bidirectional A^(*)algorithms,as well as an enhanced algorithm without obstacle factors.Evaluation metrics including path length,time,node count,and number of turns are utilized for comparison.The simulation results demonstrate that the improved A^(*)algorithm reduces the number of search nodes by 75.9%compared to the bidirectional A^(*)algorithm,decreases the number of turns by 53.6%,and accelerates runtime by 84.1%.

关 键 词：路径规划 双向A^(*)算法 引力思想 转弯代价 障碍因子 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]