• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2025, Vol. 61 ›› Issue (1): 82-91.doi: 10.3901/JME.2025.01.082

• 机器人及机构学 • 上一篇    

扫码分享

基于阻抗椭圆修正的工业机器人变导纳拖动示教方法研究

杨承霖1, 李明富1,2,3, 刘翊1,2, 雷高攀1,2   

  1. 1. 湘潭大学机械工程与力学学院 湘潭 411105;
    2. 复杂轨迹加工工艺及装备教育部工程研究中心 湘潭 411105;
    3. 焊接机器人与应用技术湖南省重点实验室 湘潭 411105
  • 收稿日期:2024-01-30 修回日期:2024-05-09 发布日期:2025-02-26
  • 作者简介:杨承霖,男,1999年出生。主要研究方向为机器人控制。E-mail:1024306302@qq.com
    李明富(通信作者),男,1979年出生,教授,博士研究生导师。主要研究方向为智能机器人、智能制造。E-mail:mingfuli@xtu.edu.cn;limingfu2001@foxmail.com
  • 基金资助:
    国家自然科学基金(52375519)和国家重点研发计划(2021YFF0500300)资助项目。

Research on Variable Admittance Drag Teaching Method of Industrial Robot Based on Damping Ellipse Correction

YANG Chenglin1, LI Mingfui1,2,3, LIU Yi1,2, LEI Gaopan1,2   

  1. 1. School of Mechanical Engineering, Xiangtan University, Xiangtan 411105;
    2. Engineering Research Center of Complex Tracks Processing Technology and Equipment of Ministry ofEducation, Xiangtan 411105;
    3. Key Laboratory of Welding Robot and Application Technology of Hunan Province, Xiangtan 411105
  • Received:2024-01-30 Revised:2024-05-09 Published:2025-02-26

摘要: 在现代工业生产的过程中,工业机器人拖动示教的研究对改善工人与机器人的交互体验、提高生产效率有很大意义。对拖动示教过程中机器人末端的运动状态进行分析,将拖动示教过程中末端的运动状态分为三个阶段以及一个急停情况,并分析各个状态下末端交互力、末端速度的变化趋势以及应该匹配的阻抗值,提出了一种同时考虑末端交互力信息、交互力增量的变导纳策略。同时针对拖动示教过程中,末端速度与目标方向不一致的问题,提出了一种阻抗椭圆转角修正算法。首先,控制机器人完成多次拖动示教任务,并记录示教过程中末端交互力、末端速度以及末端交互力与目标方向夹角;然后,基于BP神经网络表征末端交互力与目标方向夹角;最后,基于网络输出的数据修正阻抗椭圆。使得阻抗分布更加高效,提高了示教过程中机器人末端的转向效率,并通过拖动示教实验证明了算法的有效性。

关键词: 阻抗椭圆, 拖动示教, 机器人, 变导纳控制

Abstract: In the process of modern industrial production, the research of industrial robot driving instruction is of great significance to improve the interactive experience between workers and robots and improve production efficiency. The motion state of the end of the robot in the process of drag teaching is analyzed. The motion state of the end of the robot in the process of drag teaching was divided into three stages and one emergency stop, and the changing trend of the end interaction force, the end velocity and the impedance value that should be matched under each state are analyzed. A variable admittance strategy is proposed, which considered the end interaction force information and the interaction force increment. At the same time, to solve the problem that the end speed is inconsistent with the target direction in the process of drag teaching, an impedance elliptic angle correction algorithm is proposed. First, the robot is controlled to complete the teaching task of dragging several times, and the end interaction force, end speed and the angle between the end interaction force and the target direction are recorded during the teaching process. Then, the angle between the end interaction force and the target direction is characterized based on BP neural network. Finally, the damping ellipse is modified based on the network output data. It makes the impedance distribution more efficient and improves the turning efficiency of the end of the robot in the teaching process. The effectiveness of the algorithm is proved by drag teaching experiment.

Key words: damping ellipse, drag teaching, robot, variable admittance control

中图分类号: