Abstract: Aiming at the problems of poor interaction,low programming efficiency and high learning cost in the traditional teaching method of industrial robots,a new method of track demonstration and teaching of industrial robots based on hand-held teaching pen was proposed.The transformation matrix between the coordinate system and the robot base coordinate system was established by hand-eye calibration,and the calibration accuracy optimization method based on Iterative Closest Point (ICP) was proposed to improve the hand-eye calibration accuracy.Then,the pose tracking system was used to track the position and attitude of the hand-held teaching pen,and its position and attitude were transferred to the robot base coordinate system and transmitted to the industrial robot control cabinet through Ethernet,so as to control the industrial robot to reproduce the track of the hand-held teaching pen.In this paper,the robot teaching platform was built,and the industrial robot trajectory demonstration and teaching experiment was carried out.The experimental results showed that the average error and maximum error of the pose data obtained by the hand-held teaching pen were 2.19mm and 4.57mm,which could meet the teaching requirements of industrial robots.

Key words: industrial robot, trajectory teaching, robot hand-eye calibration, hand-held teaching pen

摘要： 针对工业机器人传统示教方式存在示教交互性差和编程效率低、学习成本高等问题,提出一种基于手持示教笔的工业机器人轨迹演示示教方法。该方法首先通过手眼标定建立位姿追踪系统坐标系与机器人基坐标系之间的转换矩阵,提出基于迭代最近点(ICP)的标定精度优化方法,以提高手眼标定精度。然后,使用位姿追踪系统追踪手持示教笔的位置和姿态,并将其位置和姿态转换到机器人基坐标系下,通过以太网传输给工业机器人控制柜,从而控制工业机器人复现手持示教笔的轨迹。搭建了机器人示教平台,进行了工业机器人轨迹演示示教实验。实验结果表明,利用手持示教笔得到的位姿数据平均误差为2.19 mm,最大误差为4.57 mm,能够满足工业机器人示教需求。

关键词: 工业机器人, 轨迹示教, 机器人手眼标定, 手持示教笔