Efficient Industrial Robot Calibration via a Novel Unscented Kalman Filter-Incorporated Variable Step-Size Levenberg-Marquardt Algorithm

Robots facilitate a critical category of equipment to implement intelligent production. However, due to extensively inevitable factors like structural errors and gear tolerances, the positioning error of an industrial robot is several millimeters, therefore failing to fulfill the high-precision manufacturing requirements. To address the critical problem, this work develops a novel calibration algorithm that incorporates an unscented Kalman filter and a variable step-size Levenberg-Marquardt (UKF-VSLM) algorithm for efficient industrial robot calibration with the following twofold ideas: 1) developing a novel variable step-size Levenberg-Marquardt (VSLM) algorithm to address the local optimum issues encountered by a standard Levenberg-Marquardt (LM) algorithm and 2) incorporating an unscented Kalman filter (UKF) into the proposed VSLM algorithm to suppressing the measurement noises during the calibration process. Empirical studies on a HuShu Robotics (HSR) JR680 industrial robot demonstrate that compared with state-of-the-art calibration algorithms, the calibration accuracy of the developed UKF-VSLM is 19.51% higher than that of the most accurate LM algorithm measured by the maximum error. The empirical results strongly support the superior performance of the proposed algorithm in addressing robot calibration issues.