[1]王 鑫,袁庆霓,江 涛,等.改进 RRT-人工势场法的机械臂堆垛运动方法[J].计算机技术与发展,2022,32(02):26-31.[doi:10. 3969 / j. issn. 1673-629X. 2022. 02. 004]
　WANG Xin,YUAN Qing-ni,JIANG Tao,et al.Improved RRT-Artificial Potential Field Method for Robotic Arm Stacking Movement Method[J].,2022,32(02):26-31.[doi:10. 3969 / j. issn. 1673-629X. 2022. 02. 004]
点击复制

改进 RRT-人工势场法的机械臂堆垛运动方法()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年02期
页码:
26-31
栏目:
人工智能
出版日期:
2022-02-10

文章信息/Info

Title:
Improved RRT-Artificial Potential Field Method for Robotic Arm Stacking Movement Method
文章编号:
1673-629X(2022)02-0026-06
作者:
王 鑫; 袁庆霓; 江 涛; 施辉城; 孙睿彤; 白 欢; 衣君辉
现代制造技术教育部重点实验室,贵州 贵阳 550025
Author(s):
WANG Xin; YUAN Qing-ni; JIANG Tao; SHI Hui-cheng; SUN Rui-tong; BAI Huan; YI Jun-hui
Key Laboratory of Modern Manufacturing Technology,Ministry of Education,Guizhou University,Guiyang 550025,China
关键词:
RRT; 人工势场法; 运动规划; 避障策略; 非均匀 B 样条; 机械臂
Keywords:
rapidly exploring random tree; artificial potential field method; motion planning; obstacle avoidance strategy; non-uniform B-spline; robotic arm
分类号:
TP242
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 02. 004
摘要:
快速拓展随机树算法(RRT) 在机械臂路径规划中存在随机性强、搜索效率低、规划路径长等问题,不能在货柜堆垛场景中取得相对最优的光滑路径。 对此,该文提出了一种改进 RRT-人工势场法混合算法进行货柜堆垛机械臂运动规划。 首先,对传统快速拓展随机树算法进行改进,在传统快速拓展随机树算法的全局搜索的基础上引入目标搜索,增强了随机树的搜索效率,并使用改进后的算法进行全局路径规划;其次,对人工势场法进行改进,通过使用斥力势场范围大小作为阈值修正引力函数,使用机械臂末端执行器至末位置点影响修正斥力函数,并使用改进的人工势场法对局部路径进行优化;再次,使用三次非均匀 B 样条曲线对路径进行平滑处理,并将处理后的光滑路径作为机械臂末端执行器运动的最终路径。 最后,在 Python 模拟场景中对改进算法进行可行性分析,并在 ROS 系统对机械臂堆垛运动进行应用仿真,验证了该方法的有效性和可行性。
Abstract:
The rapidly exploring random tree ( RRT ) has some problems in the path planning of the robot arm, such as strongrandomness,low search efficiency and long planning path, which cannot obtain the relatively optimal smooth path in the containerstacking scene. For this,we propose an improved RRT-artificial potential field method,a hybrid algorithm,for the motion planning of thecontainer stacking manipulator. Firstly,the traditional RRT is improved. The target search is introduced on the basis of the global searchof the traditional RRT to enhance the search efficiency of the random tree,and the improved algorithm is used for global path planning.Secondly,the artificial potential field method is improved. The gravity function is modified by using the range of repulsive potential fieldas the threshold,the repulsive force function is modified by using the influence of the manipulator end-effector to the last position point,and the local path is optimized by using the improved artificial potential field method. Thirdly,a cubic non-uniform B-spline curve isused to smooth the path, which is used as the final path of the end effector of the robot arm. Finally, the reliability analysis of theimproved algorithm is carried out in Python simulation scene,and the application simulation of the robotic arm stacking movement in theROS system verifies the effectiveness and feasibility of the proposed method.

相似文献/References:

[1]张明开 李龙澍.关于人工势场法局部极小问题的一种解决方法[J].计算机技术与发展,2007,(05):137.
　ZHANG Ming-kai,LI Long-shu.A Method for Solving Local Minimization Problem of Artificial Potential Field[J].,2007,(02):137.
[2]邢慧杰,张晓滨,张宏伟.APF与A*融合的多目标点路径规划算法[J].计算机技术与发展,2024,34(08):116.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0116]
　XING Hui-jie,ZHANG Xiao-bin,ZHANG Hong-wei.A Multi-objective Point Path Planning Algorithm Based on Fusion of APF and A*[J].,2024,34(02):116.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0116]

更新日期/Last Update: 2022-02-10