Kinematic Parameters Identification and Compensation of an Industrial Robot

Abstract

Robot kinematic calibration is of capital significance for improving the absolute positioning accuracy in many industrial applications. Aiming at identification and compensation of the kinematic errors in a 6 degree of freedom industrial robot manipulators, a technique for absolute calibration of robot is elaborated and achieved. A calibration model which takes into account all possible geometric error parameters is deduced by combining the geometrical errors in link parameters, base frame and end-effector of the robot. A minimum set of error parameters is established with 27 geometric error parameters which meets three basic requirements: model completeness, parameter minimality and model continuity. An iterative optimization process based on the least squares theory is employed in estimating the values of these kinematic errors with better precision. The effectiveness and anti-noise performance of the proposed algorithm are analyzed by the means of computer simulation. In addition, an effective compensation method based on incremental controlling algorithm is introduced which can be directly applied to Cartesian trajectory controlling of robot without having to calculate the inverse kinematics. An experimental validation with a 6 degree of freedom serial robot shows the effectiveness of the proposed method in improving the absolute positioning accuracy. Experimental results indicate that the proposed method improves the mean/maximum position errors at the measurement points on the end-effector from 2.66 mm/4.42 mm to 0.20 mm/0.36 mm respectively. Furthermore, it is obviously that the proposed technique is valid for the kinematic parameters identification and compensation of most typical 6 degree of freedom serial industrial robot [1, 2, 3, 4, 5].