一种视觉辅助机器人离线编程去毛刺系统  

A Robotic Off-line Programming Deburring System Assisted by Machine Vision

作　　者：赖增亮 刘冠峰[1] LAI Zeng-liang;LIU Guan-feng(Faculty of Electrome chanical Engineering,Guangdong University of Technology,Guangdong Guangzhou 510000,China)

机构地区：[1]广东工业大学,广东广州510006

出　　处：《机械设计与制造》2021年第9期251-254,共4页Machinery Design & Manufacture

基　　金：基于多传感器融合的多机器人协调操作理论与应用研究(2015DFA11700)。

摘　　要：目前铸造工件的自动化去毛刺作业大量使用机器人加工。大部分的解决方案是使用手工示教或者机器人离线编程,但手工示教效率较慢,而离线编程的手工标定精准度比较欠缺。由于工件的误差,使得离线编程在大批量作业中遇到大量问题。本文提出一种利用机器视觉自动检测工件轮廓的方法,解决大批量工件之间的个体误差问题,对每个工件检测轮廓后动态修改机器人运动路径。如此,在机器人自动去毛刺的作业中,使用CCD相机获取的视觉信息和自主开发的离线编程系统(RobSim)生成可执行的机器人运动轨迹。利用此方法,可有效提高去毛刺的加工质量。The automatic deburring process of casting parts has been commonly investigated by industrial robot manipulators in recent decade.Majority of the solutions are manual teaching or robot off-line programming,but manual teaching is high timeconsuming and manual calibration accuracy of off-line programming is insufficient.Due to the errors of the workpiece,offline programming encounters a lot of problems in mass production.In this paper,a method of automatic detection of workpiece contour by machine vision is proposed to eliminate individual errors between large quantities of workpieces and dynamically modify the robot motion path after detection of each workpiece contour.Thus,in the automatic deburring of the robot,the visual information obtained by CCD camera and the off-line programming system(RobSim)developed by ourselves are used to generate the executable robot movement track.In addition,it can effectively improve the processing quality of deburring with our method.

关 键 词：机器人 离线编程 机器视觉 工件去毛刺 

分 类 号：TH16[机械工程—机械制造及自动化] TP391.41[自动化与计算机技术—计算机应用技术]