基于改进VSRB-RRT算法的机器人路径规划仿真实验  

Simulation experiment of robot path planning based on improved VSRB-RRT algorithm

作　　者：倪建云[1] 李浩[1] 谷海青[1] 杜合磊 吴杰 薛晨阳 NI Jianyun;LI Hao;GU Haiqing;DU Helei;WU Jie;XUE Chenyang(School of Electronic Engineering and Automation,Tianjin University of Technology,Tianjin 300384,China)

机构地区：[1]天津理工大学电气工程与自动化学院,天津300384

出　　处：《实验技术与管理》2023年第9期172-178,共7页Experimental Technology and Management

基　　金：教育部产学合作协同育人项目(201801286006);天津理工大学教学基金项目(ZD21-06,ZD22-07GJ)。

摘　　要：针对B-RRT算法在路径规划时存在的采样效率低、路径冗长且不光滑,以及无法实时避障等问题,该文提出了基于改进的可变采样区域双向RRT(VSRB-RRT)和DWA的避障路径规划融合算法。在全局规划过程中,VSRB-RRT算法使用可变采样区域和目标偏置策略相结合的方法,加快了收敛速度并提高了采样效率,并使用贪婪优化、迭代优化和关键点优化生成代价低且光滑可执行的路径。在局部规划过程中,以改进VSRB-RRT算法规划的路径为指引,选取路径上的关键点,并使用改进DWA算法在关键点分段路径上分段规划。仿真实验表明:改进VSRB-RRT算法具有较好的搜索效率,能够以最少的时间和最稳定的效率获得最优路径,同时也验证了融合算法在实时避障路径规划时的有效性。To address the problems of low sampling efficiency,long and unsmooth paths,and the inability to avoid obstacles in real time in the B-RRT algorithm for path planning,this paper proposes a fusion algorithm for obstacle avoidance path planning based on improved variable sampling region bidirectional RRT(VSRB-RRT)and DWA.In the global planning process,the improved VSRB-RRT algorithm uses a combination of variable sampling region and target biasing strategy to speed up rate of convergence and improve sampling efficiency,and uses greedy optimization,iterative optimization and key point optimization to generate low-cost and smooth executable paths.In the local planning process,the path planned by the improved VSRB-RRT algorithm is used as a guide to select the key points on the path,and the improved DWA algorithm is used for segmented planning on the key point segmented path.Simulation experiments have shown that the improved VSRB-RRT algorithm has good search efficiency,can obtain the optimal path with the least time and the most stable efficiency,and also verifies the effectiveness of the fusion algorithm in real-time obstacle avoidance path planning.

关 键 词：路径规划 可变采样区域 改进VSRB-RRT算法 改进DWA算法 融合算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]