Font Size: a A A

Research On On-line Measurement Technology Of Assembly Pose Of Shaft Parts Based On Monocular Vision

Posted on:2022-02-05Degree:MasterType:Thesis
Country:ChinaCandidate:W X YangFull Text:PDF
GTID:2481306527495224Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Intelligent manufacturing and automatic assembly are the mainstream development direction of modern manufacturing industry.With the continuous development of industrial robot technology,assembly robot is widely used in the field of modern industry.The traditional assembly robot relies on teaching to complete the assembly work,which has low degree of intelligence and flexibility.In the process of operation,the perception ability of the production environment is weak and the flexibility is poor,and the position and pose accuracy of the workpiece assembly is easy to be disturbed by the environment.In this paper,taking the typical shaft and hole assembly as the research object,the pose positioning system of assembly robot parts based on monocular vision is studied.By analyzing the research status of vision pose measurement technology at home and abroad,based on the eye in hand vision measurement theory,the kinematics model of ER3B-C10 six axis industrial robot is deduced,and the mathematical model of automatic assembly system of shaft and hole is established.The camera calibration,TCP calibration and hand eye relationship calibration are completed by Zhang’s calibration method,six point method and Tsai Lenz algorithm,respectively.Hough’s circle changing algorithm is applied Based on canny operator,the edge information of ellipse is extracted,and an ellipse fitting algorithm based on random sampling consistency is adopted,which solves the problem of poor accuracy of least square solution of ellipse fitting when there is gross error in data,and improves the accuracy of feature extraction of circular hole and mark point Based on the principle of Pn P,Tsai two-step method is used to calculate the pose transformation matrix between the mark point and the camera.According to the coplanar constraint between the end face of the hole and the mark point,the characteristic pose calculation method of the circular hole based on the coplanar constraint is used to complete the calculation of the characteristic pose of the circular hole in the camera coordinate system.The experimental platform of shaft and hole assembly automatic assembly system is built,and the shaft and hole assembly assembly experiment is carried out.The experimental results verify the rationality and feasibility of the proposed shaft and hole assembly assembly system model.The experimental results show that the shaft and hole assembly automatic assembly system based on monocular vision has the advantages of accurate measurement results,short software development cycle and low cost,which provides a new idea for vision guided assembly robot.
Keywords/Search Tags:Monocular vision, Shaft and hole assembly, Visual positioning, System calibration, Image processing, Industrial robot
PDF Full Text Request
Related items