Research And Implementation Of Synchronous Grasping Of Workpiece On Automatic Production Line

Nowadays,as the steady development of China's industry and the booming of China's manufacturing,the automatic production line are more and more widely applied.With the development of robot technology,industrial robots are also applied in automatic production line more and more widely.For this reason,it is of great practical significance to study the synchronization grasp of workpieces by industrial robots based on industrial automation line.Based on analysis and comparison of the frequently-used methods of synchronization grasp ing workpiece in automatic production line,the thesis proposes a new methods of synchronization grasping workpiece based on machine vision,which use the image processing technology to detect and locate the workpiece's real-time coordinate in the conveyor and then set it to the industrial robot controller for controlling the robot' tracking and grabbing the workpiece.In this thesis,canny edge detection algorithm is used to detect the ROI(Region of Interest),and the randomized Hough transform is used to identify the workpiece' shape in the ROI,and then calculate the centroid of the workpiece as to obtain the two-dimensional coordinate in the image coordinate system.Then the workpiece three-dimensional coordinates the industria l robot coordinate system is obtained by transforming coordinate system.This thesis simulates the detecting and positioning algorithm on Matlab by detecting the image which containing the workpiece in the industrial conveyor.The result show that using this algorithm can obtain the actual position of the workpiece.Basic on the result of simulation,the thesis selected the suitable hardware and make up an integrated system.The image processing hardware platform of integrated system is structured by DSP chip and camera,and the data transmission tool is serial communication.To implement the image detecting and positioning algorithm on the hardware system and improve the real-time performance,this thesis simplify the algorithm and reduce the algorithm complexity.To solve the non-linear relationship between the pixel value in the original image coordinate system and the value in robot coordinate system because of the distortion in original image,use the interpolation method to establish the one-to-one correspondence between the value of y axis in the original image coordinate system and Y axial in robot's coordinate system,so that the camera can accurately determine the workpiece position.This thesis use the PI controller to reduce the error between robots' handler' position and the workpiece' position so that the robot can tracing the moving workpiece,and set the controller parameters on-site.To solve the hysteresis of robot' handler's position to the workpiece's position,this thesis use the Kalman filtering to predict next step of workpiece' position and then as the input of PI controller.This thesis build a testing system of synchronous grasping workpiece on conveyor based on machine vision on the THMSIR-1 industrial robot training platform.And the result of testing showing that the designed synchronization grasping workpiece system based on machine vision can grasp the workpiece synchronous in automatic production line.In conclusion,this thesis achieve the desired objectives of research.