城市环境下基于A*算法和DWA算法的无人机路径规划方法研究  

Research on UAV Path Planning by A*Algorithm and DWA Method in the Urban Environment

作　　者：李歆莹 房建武[1] LI Xinying;FANG Jianwu(College of Transportation Engineering,Chang'an University,Xi*an 710064,China)

机构地区：[1]长安大学运输工程学院,西安710064

出　　处：《无人系统技术》2023年第2期61-70,共10页Unmanned Systems Technology

基　　金：国家自然科学基金(62273057)。

摘　　要：为解决城市环境下无人机在全局最优路径的基础上实现动态避障的路径规划需求,提出一种将改进A*算法和动态窗口算法(DWA)结合的算法。首先,利用无人机的起点和目标点构成一个二维平面与建筑障碍物的高度关系进行坐标转化,将三维城市地图转换成二维平面地图,改善路径规划过程中耗时长和计算速度慢的问题。其次,针对传统A*算法得到的路径存在过多的转角、不利于无人机飞行的问题,利用弗洛伊德(Floyd)法对全局路径进行平滑处理。最后,提取平滑处理后的路径父节点作为DWA算法的中间目标点,实现局部规划。实验结果表明,在不同类型的环境下,提出的融合算法既能保证更平滑和距离更短的全局最优路径,又能有效地躲避环境中出现的动态障碍物。To meet the requirement of dynamic obstacle avoidance path planning for a UAV based on the global optimal path in the urban environment,an algorithm combining an improved A*algorithm and a dynamic window algorithm(DWA)is proposed.First,to solve problems of time-consuming and low calculation speed in the process of 3D environment path planning,converting the 3D map into a 2D grid map by coordinate conversion which is based on the relationship between the plane formed by the starting point and the target point and the height of obstacles.Second,aiming at the problem that the path obtained by the traditional A*algorithm has excessively large turning angles,which is not conducive to the flight of the UAV,the Floyd algorithm is used to smooth the global path.Then,extract parent nodes of the smoothed path as the intermediate target points in the DWA algorithm to achieve local path planning.The experimental results show that in different environments,the proposed fusion algorithm can not only guarantee a smoother and shorter global optimal path but also effectively avoid dynamic obstacles around.

关 键 词：路径规划 A*算法 动态窗口算法 动态避障 环境建模 栅格法 

分 类 号：TP249[自动化与计算机技术—检测技术与自动化装置]