FANUC Robot Grasping System Based On Binocular Structured Light

Driven by rising labor costs and the rapid development of science and technology,the logistics industry,various manufacturing industries and high-risk industries are all undergoing automation transformation,and the concept of machine substitution has been proposed.Industrial robots have been widely used in modern industry.In many factories,industrial robots are applied to the positioning and grasping of workpieces,but the workpieces are often scattered and disorderly.2D vision technology can only grasp neatly placed workpieces,which is less practical than 3D technology.In view of the shortcomings of the current 3D reconstruction technology in my country and the application scenarios of actual needs,combined with domestic and foreign references.This thesis takes the FANUC six-axis industrial robot as the research object,and combines the binocular structured light vision technology to carry out three-dimensional reconstruction and positioning and grasping of the disorderly placed workpiece.Different from the previous monocular cameras that need to calibrate the projector,this thesis calibrates the camera and robot.The robot coordinate system is calibrated by using the calibration plate as an auxiliary coordinate system combined with the projection formula of the vector.In order to reduce the amount of calculation when the corresponding points are matched,the corresponding points are calculated by combining the epipolar constraint and the uniqueness of the Gray code value.In point cloud segmentation,RANSAC is combined with cluster segmentation method and region growing method for experimental comparison,and a better method is selected.Use ROBOGUIDE to compile the robot control program,write the QT human-computer interaction interface,and realize the automatic grasping of the six-axis industrial robot.The main content of this dissertation is divided into five sections: 1.The overall design of the three-dimensional positioning and grasping system of the workpiece according to the actual needs;2.The design and construction of the system hardware type and experimental platform are completed;3.The calibration of the binocular camera,Structured light coding and processing,corresponding point matching,point cloud processing,and establishment of3 D models;4.Kinematics analysis of industrial robots;5.Testing system performance using experimental platforms.