无人机路径规划源于机器人运动规划，是当下无人机应用研究的核心内容，对提高无人机系统在复杂环境中的作业能力起着关键作用。针对快速扩展随机树(Rapidly-exploring Random Tree，RRT)算法进行无人机路径规划时搜索随机性高、存在冗余路径和路径平滑性差的问题，提出了一种面向无人机路径规划的改进RRT算法。改进RRT算法在RRT算法的基础上结合人工势场法中的引力函数使得随机节点的产生具有目标导向性，限制了随机树的拓展方向，从而降低了搜索的随机性;结合贪心算法对规划所得路径进行剪枝优化，去除冗余节点，缩短了路径长度;结合B样条曲线对路径进行平滑性处理，去除曲率突变的转折点，形成一条平滑的适合无人机实际飞行的路径。通过仿真软件对A*算法、传统RRT算法与改进RRT算法进行对比分析，仿真结果表明，提出的改进 RRT算法性能更高，在狭窄通道场景与复杂障碍物场景下相比于传统RRT算法平均规划时间各减少 49。44%和 17。97%，相比于 A*算法平均规划时间各减少了80。23%和52。93%，得到的路径更短更为平缓，同时大幅降低了RRT算法路径规划失败的可能性，验证了改进RRT算法的可行性与有效性，解决了原算法随机性高、存在冗余路径和平滑性差的问题。
Unmanned Aerial Vehicle Path Planning Based on Improved RRT Algorithm
Unmanned aerial vehicle path planning,derived from robot motion planning,is currently a central focus in unmanned aerial vehicle application research and plays a crucial role in enhancing the operational capabilities of unmanned aerial vehicle systems in complex environments.In this paper,aiming at the problems of Rapidly-exploring Random Tree(RRT)algorithm in unmanned aerial vehicle path planning,such as high search randomness in search,redundant paths and poor path smoothness,an improved RRT algorithm for Unmanned Aerial Vehicle path planning is proposed.The improved RRT algorithm combines the gravitational function from the artificial potential field method with the RRT algorithm to guide the generation of random nodes in a directed manner,thereby con-straining the expansion direction of the random tree and reducing search randomness.Secondly,the greedy algorithm is combined to prune and optimize the planned path,shortening path length.Finally,the B-spline curve is combined to smooth the path,removing the turning points of curvature mutations,and forming a smooth path suitable for the actual flight of unmanned aerial vehicles.A comparison and analysis of the A* algorithm,the traditional RRT algorithm and the improved RRT algorithm are conducted using simulation soft-ware.Simulation results show that the improved RRT algorithm proposed in this paper has higher performance,with average planning time improvements of 49.44%and 17.97%compared to the traditional RRT algorithm in narrow-channel scenarios and complex obsta-cle scenarios,and 80.23%and 52.93%compared to the A*algorithm.The generated paths are shorter and smoother,while significant-ly reducing the possibility of path planning failures compared to the RRT algorithm.This verifies the feasibility and effectiveness of the improved RRT algorithm,addressing the issues of high search randomness,redundant paths and poor path smoothness in the original al-gorithm.
unmanned aerial vehiclepath planningimproved RRT algorithmRRT algorithmA* algorithmgravitational function
国家科技期刊平台
NSTL
万方数据
