The Modeling And Parameter Identification Method For Calibration Of Industrial Robot

With the widespread of the domestic industrial robot application and the need of complex task, the demand of robot accuracy becames higher and higher. The robot calibration can compensate the distortions caused by several factors with minimum, and improve the absolute positioning accuracy of robot, and instead of changing the robot structure to improve the performance of the transmission mechanism. This paper aims at a future study on the modeling method of robot calibration technology and the parameter identification methods, to complete industrial robot calibration tasks rapidly and accurately, consequently to improve the absolute position accuracy of industrial robots.According to the research literature, 95% of the robot positioning error comes from the inaccuracy of kinematic model. Most of the current robot calibration model are based on the 4DOF parameter DH model, while the most significant drawbacks of this model are the kinematic parameter sigularity and the problemof sudden chang. Focus on this situation, this paper established the model of MDH which has five degrees of freedom, and then established the model of CPC as well as the model of MCPC which have 6 degrees of freedom. These two models have the completeness and continuity which can accomplish posture error calibration compensation. Combining the robotic toolbox with MATLAB GUI, using the numerical and graphical method,the correctnessof these four models are verified.Combining kinematics model with differential transformation method, the jacobian matrix of the robot is derived, as well as the position of the robot calibration error model, the relative distance error model and the constraint of minimum error model is established.With the distance error model derived from the differential transformation, a relative distance’s square error model is established.Focusing on the OTC arc welding robot, this paper analyzes the robot error sources. Based on the four kinematics models, this paper analyzes the sensitivity and the influence law of the various parameters to the endeffector of the robot. This analysis provides athoritical foundation to the industrial robot parts processing, distribution and the parameters of the calibration precision compensation.Because of the redundancy of the parameters, the error model used in calibration will seriously affect the accuracy and robustness in the kinematic parameter identification. To this remedy the situation, the paper discussed the necessity of the redundancy of the parameters in theory firstly. Then based on the calibration models of DH and MCPC, the paper analyzes the adjacent joint correlation coefficient matrix through the jacobian matrix is derivation. Therefore, the position error model of two kinematic calibration under different identification conditions and strycture of serial robots,thedistance error model and the redundancy parameters of the distance square error model are derived theoritically. The paper had done the redundancy of the parameters of the simulation experiment of the six degrees of freedom of OTC arc welding robot.Basing on the parameter identification algorithm of the least squares method, optimization algorithm, genetic algorithm and simulated annealing algorithm, the paper designs the GUI of calibration,and sets up calibration experiment platform, for the aquirtion of data. Combining with six degrees of freedom of OTC arc welding robot calibration results, the paper conclude the general theoritical law.