The Industrial Robot Geometry Parameters Calibration Based On Vision

With the rapid development of the manufacturing industry,the application of industrial robots has become more and more extensive,especially in the production line.In order to allow the robot to accurately capture the workpiece on the production line,it is necessary to assemble the visual equipment to cooperate with the robot.However,due to the machining and assembly errors,there is a deviation between the actual and the theoretical model parameters of the robot.Therefore,the geometric error of the robot body needs to be calibrated.After calibrating the parameters of the robot model,it is necessary to calibrate the hand-eye relationship between the vision device and the robot base.At present,the commonly used robot body calibration devices include a laser instrument and a cable puller,but these devices have high cost and complicated calibration process,and the visual device has the characteristics of low cost,simple installation and non-contact measurement.Therefore,this paper selects the visual method to calibrate the geometric parameters of FD six-degree-of-freedom industrial robot.The details are as follows:Firstly,a high-precision industrial camera is installed at the end of the robot,and moving calibration plate to collect multiple images,the Matlab calibration toolbox is used to calibrate the camera internal parameters,and the camera external parameters are obtained by Zhang's calibration method,then the position accuracy of the camera is verified according to the distance error model.Secondly,the DH modeling method is used to establish the FD robot kinematics model.The position error model of the robot is obtained by kinematic differentiation.The distance error model is used to construct the equation about the geometric error of the robot body.The robot is controlled at different positions and postures according to the planned trajectory.The calibration plate image is taken and the camera position is calibrated,and then the geometric parameter error is identified multiple times using the least squares method to make the actual model of the robot close to the theoretical.Thirdly,conduct a study of the hand-eye calibration.The first step is to plan the circular motion trajectory of the calibration plate,and the position translation vector between the calibration plate in the specified attitude and the trajectory center point is calibrated.The second step is to construct the calibration equation for the hand-eye geometric parameters,and the third step is to use the pseudoinverse method to solve the equations.Finally,conduct an experimental analysis and summary,and propose the next step for improvement.