Font Size: a A A

Research On Disordered Valve Placement Grabbing System Based On Machine Vision

Posted on:2022-02-01Degree:MasterType:Thesis
Country:ChinaCandidate:W H ZhangFull Text:PDF
GTID:2492306326466864Subject:Master of Engineering
Abstract/Summary:PDF Full Text Request
The traditional manufacturing industry will be equipped with industrial robot system to complete loading and unloading operations,but the industrial robot can only be programmed in a solidified way to complete a specific task,the need for manual cooperation to put the material in a specific position at a specific Angle.For grasping the valve,because of the small size of the valve,the circular end surface features are not obvious,and there is no fixed rule in the material frame,the simple teaching programming method can not be used to control the robot to achieve the purpose of fast automatic grasping the valve.Therefore,in order to improve the degree of automation of industrial robots and improve the efficiency of valve grasping work,this paper designed a disorderly placement grasping system for valve workpiece based on the reference of a large number of literature and machine vision system application examples.OpenCV was used to calibrate the camera,and Kinect depth camera was used to collect the image of the valve workpiece.Then,the image was recognized and processed to solve the pose of the target workpiece.The experimental platform of the valve disordered placement grasping system was built,and each module of the system was designed in blocks.Then a variety of strategies are used to simulate the valve disorderly placement grasping system to verify its practical significance and value.The main contents of this paper are as follows:The image collected by Kinect depth camera was recognized and positioned,and the geometric features of the valve workpiece in the image were analyzed.Firstly,the color image is processed by grayscale and binarization,and then the edge is extracted by Sobel operator.The improved marker watershed algorithm based on morphological reconstruction is used to segment the image,and the Hough transform based on chain code algorithm is used to fit the contour of the 2D preprocessed image,and then the template matching is carried out.Machine vision system calibration.By analyzing and modeling the disordered valve placement grasping system,image coordinate system,camera coordinate system,world coordinate system and manipulator coordinate system are constructed respectively.In order to solve the transformation relationship between each coordinate system,OpenCV software is used to calibrate the camera,determine the relationship between the world coordinate system,the camera coordinate system and the image coordinate system,and perform eye-in-hand hand-eye calibration to determine the relationship between the camera and the robotic arm coordinate system.The relationship between the image coordinate system and the manipulator coordinate system can be obtained,and then the three-dimensional pose of the target workpiece in the manipulator coordinate system can be obtained.Simulation test valve disorderly placement grasping system.Using MATLAB and C++ mixed programming method,the robot and the visual system Socket communication,complete the transmission of the workpiece pose data.The experimental platform of valve disorderly placement grasping system was built to test the grasping valve workpiece and verify the feasibility,reliability and accuracy of the system.
Keywords/Search Tags:Machine vision, Image processing, Industrial robot, System calibration, Vision positioning
PDF Full Text Request
Related items