基于融合改进RRT^(*)算法与改进DWA算法的路径规划研究  

Research on path planning based on merging improved RRT^(*) algorithm and improved DWA algorithm

作　　者：吴天文 李洪兵[2] 袁浩 曾蝶 王语凝 Wu Tianwen;Li Hongbing;Yuan Hao;Zeng Die;Wang Yuning(School of Electronic and Information Engineering,Chongqing Three Gorges University,Chongqing 404120,China;School of Computer Science and Engineering,Chongqing Three Gorges University,Chongqing 404120,China)

机构地区：[1]重庆三峡学院电子与信息工程学院,重庆404120 [2]重庆三峡学院计算机科学与工程学院,重庆404120

出　　处：《国外电子测量技术》2023年第5期77-88,共12页Foreign Electronic Measurement Technology

基　　金：国家重点研发计划项目(2021YFB3901405);重庆市自然科学基金(2022NSCQ-MSX4084);重庆市教委科学技术研究项目(KJZD-M202201204,KJQN202101233,KJQN202001229);重庆市人工智能+智慧农业学科群开放基金(ZNNYKFB201901);重庆市三峡库区地质环境监测与灾害预警重点实验室开放基金(MP2020B0202)项目资助。

摘　　要：针对传统动态窗口算法(dynamic window approach, DWA)在动态环境下预测能力较差且非最优路径等问题,提出一种改进的DWA算法。通过密度约束和采样区域限制来改进RRT^(*)(rapidly-exploring random tree)算法,提高RRT^(*)算法的收敛速度。将改进RRT^(*)算法得到最优路径作为机器人移动的参考路线,以保证全局的路径最优。通过预测轨迹的末端状态来合理的延长预测轨迹,以传递影响到评价函数,使机器人可以提前探测到障碍物并进行避障动作。在不同环境下,改进后的算法移动路径长度和消耗时间分别减少12.76%和30.14%,且具有更强的前瞻性以及更短的路径。Aiming at the problems of poor prediction ability and non-optimal path of traditional dynamic window algorithm dynamic window approach(DWA)in dynamic environment,an improved DWA algorithm is proposed.Rapidly-exploring random tree(RRT^(*)) is improved by density constraint and sampling area limitation,and the convergence speed of the RRT^(*) algorithm is improved.The optimal path obtained by improving the RRT^(*) algorithm is used as the reference route for robot movement to ensure the global path optimization.By predicting the end state of the trajectory,the predicted trajectory is reasonably extended to transmit the impact evaluation function,so that the robot can detect obstacles in advance and perform obstacle avoidance actions.Under different environments,the improved algorithm reduces the moving path length and consumption time by 12.76%and 30.14%,respectively,and has a stronger forward-looking and shorter path.

关 键 词：RRT^(*)算法 DWA算法 动态环境 预测轨迹 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]