基于改进RRT的驼峰自动摘钩机器人路径规划算法  

Path Planning Algorithm for Automatic Hump Uncoupling Robot Based on Improved RRT

作　　者：栾德杰 冯军 杨华昌 曹子昱 张辉 LUAN Dejie;FENG Jun;YANG Huachang;CAO Ziyu;ZHANG Hui(Signal&Communication Research Institute,China Academy of Railway Sciences Corporation Limited,Beijing 100081,China)

机构地区：[1]中国铁道科学研究院集团有限公司通信信号研究所,北京100081

出　　处：《铁道运输与经济》2023年第1期30-38,46,共10页Railway Transport and Economy

基　　金：中国铁道科学研究院集团有限公司科研项目(2021YJ021)。

摘　　要：为实现编组站驼峰摘钩技术机械化、自动化和智能化,以自动摘钩机器人路径规划算法为研究重点,提出了一种基于RRT的驼峰自动摘钩机器人路径规划算法。针对驼峰摘钩作业特点对驼峰自动摘钩机器人路径规划问题采用3种特征描述,并结合作业过程对路径规划采样范围进行限定,增强算法的目标导向性;在RRT的基础上利用人工势场法引入目标引力和障碍物斥力,提高路径搜索的方向性以加快路径搜索速度;通过三阶贝塞尔曲线对搜索到的路径进行平滑处理,确保路径的可行性。仿真结果表明,改进后的算法具有运算速度快、规划能力强、稳定性高的特点,能够根据摘钩机器人作业特点完成路径规划,进而解决摘钩机器人机械臂作业时关节运动和障碍躲避等问题。This paper focused on the path planning algorithm for automatic uncoupling robots and proposed a path planning algorithm based on the rapid-exploring random tree(RRT) for automatic hump uncoupling robots. The effort aimed to realize the mechanical, automatic, and intelligent hump uncoupling technology in the marshaling station. Firstly,three feature descriptions were adopted for the path planning of the automatic hump uncoupling robot in view of the hump uncoupling features, and the sampling range of path planning was limited considering the operation process to enhance the goal orientation. Secondly, the artificial potential field method based on RRT was used to introduce the target gravity and obstacle repulsion to improve the direction of path search and speed up path search. Finally, the searched path was smoothed through the third-order Bezier curve to ensure the feasibility of the planned path. The simulations show that the improved algorithm has the characteristics of fast operation speed, strong planning ability, and high stability,which can plan paths according to the operation characteristics of the uncoupling robot and then solve the problems of joint motion and obstacle avoidance during the operation of the robot arm.

关 键 词：编组站驼峰 自动摘钩机器人 RRT算法 路径规划 人工势场 

分 类 号：TP242.2[自动化与计算机技术—检测技术与自动化装置]