Research On Key Technology Of Material Sorting Industrial Robot Based On Machine Vision

With the extensive application of industrial robot technology in the manufacturing field,more and more industrial robots are applied to manufacturing industry,replace the manual to complete the sorting of workpiece on the production line.Although the industrial robot can complete the sorting task by teaching programming,the surrounding equipment need to match the sorting system,and the cost is also high.For multiple targets,the shape,size and the position of the workpiece are not sure of the sorting task,how to make the robot recognize and locate the workpiece,and grasp the target accurately is the key to make the sorting robot more intelligent.The robot has the function of human eyes by applying the machine vision technology to the sorting robot,and it is great significance for industrial robots to achieve the goal of grasping the workpiece by visual guidance.In this paper,the technology of KR 6 R700 sixx industrial robot for visual sorting is studied.A single industrial robot can not locate the position of the material on the working plane.In order to enable the robot to recognize the location of the workpiece,the method of visual recognition and location based on image sensor is designed,and the positioning of machine vision is achieved.The positioning technology of machine vision is applied in KUKA industrial robot to make the robot determine the position of target material.The coordinate frames of links is established,then D-H parameters is determined.The kinematics equation of the robot is established,and the positive solutions and inverse solution of the KUKA industrial robot is obtained.The trajectory planning in joint space of KUKA robot is studied using five polynomial curve.The workspace of the robot is analyzed on the MATLAB software platform using the forward kinematics solution.Industrial camera serve as the image sensor for sorting robot.Four coordinate systems in vision system are analyzed,and the linear imaging model and nonlinear imaging model of industrial camera are analyzed.The method of calibration for industrial camera based on planar target is studied,and the calibration experiment of industrial camera is constructed,then the internal and external parameter matrix of industrial camera is obtained.The visual positioning experiment for the 18cm×18cm region on the sorting work plane is implemented,and the positioning error is analyzed.The maximum positioning error is less then 0.9mm.After industrial camera captured images of workpiece,the bilateral filter is used to filter the image.The binary image of workpiece is obtained using the Otsu threshold segmentation method.The contour of the workpiece is extracted,and the identification of the workpiece is completed by judging the area of the contour.The pixel position of the workpiece is obtained by extracting the center of the contour.The angle calculation method is designed to obtain the angle information of the workpiece.The pose data communication between the PC and KUKA industrial robot is analyzed,and the communication format of pose data for KUKA robot is determined.The PC server to send data is built,and the communication between PC and robot is established.Finally,the visual sorting platform is built and the sorting experiment is carried out.The result show that the industrial robot sorting system can complete the sorting of the target workpiece.The adaptability of the sorting robot is improved.