基于地图分解AGV全局路径规划新方法
打开文本图片集
关键词：地图分解；A*算法；麦克纳姆轮；路径优化；全局路径规划 中图分类号：TP242 文献标志码：A 文章编号：1001-3695（2025）08-015-2355-09 doi： 10.19734/j.issn.1001-3695.2024.12.0527
New approach to global path planning for AGV based on map decomposition
Wang Hanga†，Huang Xixiab，Liang Donga （a.InstituteofLgticsSiee&Engneing，bKeyLbatoryofspoIdustryfrineecholog&ontrolEngeringh hai Maritime University，Shanghai ，China）
Abstract：Toaddress the issues of excessivenode traversal，poorpath smoohness，and longsearch time intraditional A* algorithmsforautomated guidedvehicle（AGV）path planning inlarge-scalescenarios，thispaperproposedathree-layerblock search A* algorithm（Blocks- ⋅A∗ ）and developed a Mecanum-wheel AGV to resolve kinematic constraints.The Blocks-A* algorithmdecomposedthemapinto largerregions（blocks），replacing node-level searches with block-leveloperations.Firstly， it divided free spaceandrestricted space basedon priormap information，partitions free space intotriangles，performed node equivalence for these triangles，and constructed adjacency matrices. Secondly，the Blocks-A algorithm calculated the optimal blockchanneltogeneratesuboptimalpathsusingmidpointsofadjacentrangleboundaries.Thirdly，itappliedlinearprogamming or quadraticprogrammingoptimizationmodels according to specificscenarios togenerate thefinaloptimalpath.Experimentalesultsdemonstratethattheproposedalgorithmsignificantlyreduces traversednodes inlarge-scaleenvironments，improves pathsmoothnesstobetter meetAGVmotionrequirements，and enhances search eficiency.This method isspecifically applicable to large-scale path planning problems.
Key words：map decomposition；A* algorithm； Mecanum-wheel；path optimization； global path planning
0 引言
AGV作为智能物流和智能制造系统中的核心组成部分，近年来在多样化应用场景和复杂实际需求驱动下得到了广泛部署和发展，被应用于超大型电商与仓储物流中心、国际港口自动化、汽车制造超级工厂和航空枢纽等大型场景下货物运输，例如京东“亚洲一号"智能物流园，Tesla超级工厂“无轨化生产线”以及上海洋山港码头[1]。（剩余19923字）