System And Application Software Development For Intelligent Robot Of Handling And Assembly Based On Vision Guidance

With the continuous development of technology,machine vision has become one of the core technologies of robot intelligence perception.Machine vision is applied in the field of industrial robots,which is dramatically significant to improve the intelligent and adaptability of robots in industrial production.In this dissertation,the design and development of the intelligent robot for the handling and assembling based on vision guidance are systematically presented according to the requirements of the handling and assembly of typical workpieces.Firstly,the related research and development of the handling and assembly robot based on vision guidance at home and abroad is elaborated.Based on the detailed analysis of the requirements of heavy duty wheel handling and assembly,the system structure is designed aiming at robot controller with different feature of openness and expansion,and the design of the vision guidance system based on monocular camera "eye-in-hand" and "eye-to-hand" is presented according to the working principle of the vision system in different applications.Aiming at the calibration of the vision system which directly affects the positioning accuracy,the imaging model,the camera model and the hand-eye calibration principle are discussed.Besides,the procedure of the camera calibration and hand-eye calibration are given in detail.To realize the fast recognition,positioning,grasping and placing technology,the method based on the characteristics of the contour,the fitting hole and the SIFT local feature are discoursed,and the transformation of coordinate from image coordinate system to world corrdinate system is implied.Combined with the control requirements of handling and assembly,the design of assembly process,the planning of the workpiece gripping position and the gripping path are designed in detail.On the basis of summarying the error source in workpiece gripping,an hole-matching algorithm based on grasping deviation compensation is proposed.In addition,vision operation control software based on OpenCV and Qt is developed independently,with the design and implementation of camera acquisition,visual calibration,communication and guidance control discribled in detail.Finally,the KUKA KR16 industrial robot is used as the system verification platform with the performance test of the vision guidance system and evaluations given for the functions and effectiveness of the vision guidance system.