基于凸角点搜索的自适应FMT^(*)路径规划  

Convex Corner Search-Based Adaptive FMT^(*)PATH PLanning

作　　者：原慧琳 陈昊文[1,2] YUAN Huilin;CHEN Haowen(School of Management,Northeastern University at Qinhuangdao,Qinhuangdao 066004,China;College of Information Science and Engineering,Northeastern University,Shenyang 110819,China)

机构地区：[1]东北大学秦皇岛分校管理学院,秦皇岛066004 [2]东北大学信息科学与工程学院,沈阳110819

出　　处：《组合机床与自动化加工技术》2025年第3期5-9,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：东北大学资助产学研战略合作项目(71971050);东北大学秦皇岛分校产学研基地合作战略框架协议项目(7043902891801)。

摘　　要：针对快速进行树算法(FMT^(*))存在的冗余搜索和初始采样点数量难以确定的问题,在FMT^(*)算法基础上提出了一种基于凸角点搜索的自适应快速进行树移动机器人路径规划算法。设计了一种快速检测障碍物边缘凸角点的检测器和一种懒惰邻居判断法,根据凸角点数量和地图尺寸自适应确定初始采样点数量,通过模拟不同精度地图的方式设计了等距采样并重新设计了采样点分布不均匀的搜索半径。最后使用不同类型地图进行仿真实验,仿真结果表明所提出的算法相较于传统FMT^(*)算法在取得质量更好路径的前提下搜索效率也更高。Aiming at the problems of redundant search and difficult to determine the number of initial sampling points in fast marching tree algorithm(FMT^(*)),an adaptive fast marching tree mobile robot path planning algorithm based on convex corner search is proposed on the basis of FMT^(*).A fast detector for detecting the convex corner points on the edge of obstacles and a lazy neighbor judgment method are designed.The number of initial sampling points is adaptively determined according to the number of convex corner points and the size of the map.Equidistant sampling is designed by simulating different precision maps,and the search radius of uneven distribution of sampling points is redesigned.Finally,different types of maps are used for simulation experiments,and the simulation results show that the proposed algorithm is more efficient than the traditional FMT^(*)algorithm on the premise of obtaining better quality paths.

关 键 词：移动机器人 路径规划 快速进行树 凸角点 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]