Research On Automatic Tool-Setting System Of Hybrid Robot Based On Binocular Vision

With the rapid development of industrial automation,the machine vision is applied more and more widely,while enhancing the intelligence of the machine greatly.In order to solve the problem that the process of tool-setting of engrave robot is still manual,improve the efficiency and accuracy of tool-setting and reduce it's risk,this thesis developed a automatic tool-setting system that based on binocular vision,in order to achieve the process of non-contact automatic tool-setting.Introduced the machine vision into the process of tool-setting,enhance the intelligence of engrave robot.This has important theoretical significance and practical value.This thesis has carried on the thorough analysis and the research in the following several aspects.Firstly,based on understanding of the development history of hybrid robots and machine vision,as well as the current research status and current application fields at home and abroad,the research direction of the subject was clarified and the research program was confirmed.In view of the 3TPS/TP hybrid engraving robot,based on the principle of machine vision,the research program of automatic tool-setting system of engraving machine was drawn up.Secondly,starting from the basic principle of binocular vision,using the principle of pinhole imaging and Zhang Zhengyou's plane labeling method(referred to as "Zhang's" calibration method),the calibration system mathematical model of camera's internal and external parameters was established,the camera lens distortion was nonlinearly optimized;Then,based on Matlab,the camera's internal and external parameters by experiment were obtained.Thirdly,in order to obtain the pixel coordinates of the target point,the image edge detection algorithm was chose based on Canny operator by comparison.The coordinate transformation matrix was established and the 3D reconstruction of the target point was done.Thereby obtained the 3D coordinates of the target point in the world coordinate system.Though the 3D coordinates of the initial point of machining and the initial point of the tool,the movement path required by automatic tool-setting can be further calculated.Then,an automatic tool-setting program with human-computer interaction function was developed.The system can achieve the target image acquisition,the calibration of camera internal and external parameters,the extraction of the pixel coordinates of the target point,and the 3D coordinates of the target point and the calculation of the tool-setting path.It can also output the G code which could be recognized by robot.Achieved the function of automatic tool-setting process based on binocular vision.Finally,the system has been experimentally verified.The experimental results showed that the precision of the system was less than 3mm and the relative error was less than 1.7%.The time of tool-setting was shortened by 50%compared with that of manual setting.The time of single tool-setting was 2.5min and the time of repetitive setting was 30s.Improve work efficiency and precision of the tool-setting.