基于动态运动基元的机器人电子元器件装配轨迹规划方法研究
李成，周紫菱，汤卿*
（四川大学 机械工程学院，四川 成都 610065）
摘要：针对电子元器件装配任务中电子元器件和PCB板种类较多，以及生产过程中产品种类更换频繁所导致的装配轨迹多变等问题，提出一种基于动态运动基元的机器人轨迹规划方法。首先，建立动态运动基元系统方程，对机器人的运动进行建模；其次，根据演示轨迹对非线性函数的参数进行学习，利用非线性最优化的局部加权回归方法求得演示轨迹的权重系数；然后，对方程中的非线性函数项进行分段加权，使系统能够稳定收敛到目标装配点；最后，设计具有末端姿态约束的机器人拖动算法在Panda机器人上进行轨迹演示，根据目标位姿不发生变化与发生变化两种情况生成机器人的自主装配轨迹，并在Panda机器人上对三种不同的电子元器件进行装配验证。结果表明：基于改进后的动态运动基元方法学习到的机器人运动轨迹，不仅能够很好地模仿演示轨迹的运动趋势，而且可以以较高的精度到达目标指定位姿，很好地完成了机器人电子元器件装配任务。
关键词：装配；动态运动基元；轨迹规划
中图分类号：TP242.6 文献标志码：A doi：10.3969/j.issn.1006-0316.2023.01.010
文章编号：1006-0316 (2023) 01-0059-09
Research on Assembly Trajectory Planning Method of Robot Electronic Components Based on Dynamic Movement Primitives
LI Cheng，ZHOU Ziling，TANG Qing
( School of Mechanical Engineering, Sichuan University, Chengdu 610065, China )
Abstract：Considering that the variety of electronic components and PCB boards in the electronic component assembly task and the frequent replacement of product types in the production process leads to the change of assembly trajectories and other problems, in this paper, a robot trajectory planning method based on dynamic movement primitives for electronic component assembly is proposed. Firstly, the system equation of dynamic movement primitive is established to model the robot motion; secondly, the parameters of the nonlinear function are learned according to the demonstration trajectory, and the weight coefficient of the demonstration trajectory is obtained by using the local weighted regression method of nonlinear optimization; then, the nonlinear function term in the equation is weighted piecewise, so that the system can stably converge to the target assembly pose; finally, a robot dragging algorithm with end pose constraints is designed to demonstrate the trajectory on the Panda robot. The autonomous assembly trajectory of the robot is generated respectively when the target pose changes or do not change, and the assembly of three different electronic components is verified on the Panda robot. The results show that the robot motion trajectory learned based on the improved dynamic movement primitive method can not only imitate the motion trend of the demonstration trajectory well, but also reach the target specified pose with high precision, which completes the assembly task well.
Key words：assembly；dynamic movement primitives；trajectory planning
———————————————
收稿日期：2022-06-17
基金项目：四川省科技计划（2020YFG0116，2020YFG0074）
作者简介：李成（1998－），男，四川达州人，硕士研究生，主要研究方向为机器人运动控制、人机协作，E-mail：15760469379@163.com。*通讯作者：汤卿（1982－），男，四川成都人，博士，副教授、硕士生导师，主要研究方向为机器人的设计与制造、机器人感知控制与规划、人机协作与共融，E-mail：tangqing_scu@163.com。
设为首页 | 加入收藏 | 免责条款
《机械》杂志版权所有 Copyright©2008-2012 Jixiezazhi.com All Rights Reserved
电话：028-85925070 传真：028-85925073 E-mail：jixie@vip.163.com
地址：四川省成都锦江工业开发区墨香路48号 邮编：610063
Powered by PageAdmin CMS