Research On Kinematic Calibration Of Industrial Robot Based On MDH Model

With the application of industrial robots becoming more and more widely,people have put forward higher requirements for the accuracy of industrial robots.The repetitive positioning accuracy of industrial robots is high,but their absolute localization accuracy is low,which limits their application in some fields.At present robot calibration technology is the most effective way to improve the absolute positioning accuracy of industrial robots.So in this paper,the kinematic calibration of industrial robots is studied to improve the absolute positioning accuracy of industrial robots.This paper takes IRB4600 industrial robot as the object of study.Firstly,according to the D-H modeling method,the kinematic model of industrial robot is established and the derivation of forward kinematics and inverse kinematics is completed.Then the kinematic model is simulated and verified by using Robotics toolbox.The factors that affect the accuracy of robot end localization are analyzed.In view of the singularity and parameter variation of the D-H model at the parallel axis,this paper uses the MDH modeling method to model the parallel axis.Then,the localization error model of the robot is derived based on the idea of differential transformation.The least squares method is used to identify the parameters and the robot is compensated by the method of pre compensation.The redundant parameters in the error model have a great influence on the accuracy of parameter identification.In view of this situation,the influence of redundancy of kinematic parameters on the accuracy of parameter identification is analyzed theoretically.Then by using the method of error analysis proposed by Khalil,based on the deduction of geometric relations of adjacent joints,the redundant parameters of the position error model of serial robots with different structures are analyzed theoretically.Taking IRB4600 industrial robot as the object,the analysis of redundant parameters is verified by simulation.The calibration data are collected in the whole working space of the robot,and the robot calibration experiment is carried out.the experimental results show that the robot absolute positioning accuracy is improved by 89.75% and the robot absolute positioning error is less than 0.5mm.By comparing the error compensation results ofincluding redundant parameters and removing redundant parameters in the collection points,the correctness of the redundant parameter analysis is verified.In addition,the calibration experiment of the local workspace was carried out.The test points were collected in the local working space and out of the local work space to compare the error compensation effect of the two calibration.The experiment results show that the absolute positioning accuracy of the robot in the calibration space can be further improved by reducing the calibration space.