Simultaneous compensation of geometric and compliance errors for robotics with consideration of variable payload effects

• Published in: Mechatronics (Oxford) Vol. 102; p. 103228
• Main Authors: Li, Hung-Ming, Liu, Chien-Kuan, Yang, Yong-Chun, Tsai, Meng-Shiun
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2024
• Subjects: Compensation; Compliance error; External payload; Variable payload; External payload; Variable payload; Compliance error; Compensation
• ISSN: 0957-4158
• DOI: 10.1016/j.mechatronics.2024.103228

•The core findings and main contributions can be summarized by the following points.•Proposed a novel variable payload method for parameter identification and precision compensation.•Simultaneous identification of kinematic parameter errors and compliance coefficients.•Compensate for geometric errors and compliance errors caused by joint flexibility.•Achieve higher accuracy under different payloads compared to the conventional kinematic algorithm. In order to satisfy the high accuracy requirements of robotic applications, it is necessary to consider not only the geometric errors but also the compliance errors which are caused by the self-gravity of the link and the external payloads. A general error model is developed based on the modified Denavit-Hartenberg (DH) model. For the parameters in the error model, there is a coupling between the compliance coefficients and the link parameters, making it difficult to use only the compliance coefficients to compute the compliance errors due to external payloads. As the external payload for the robot manipulator varies, the parameter identification of the model should be conducted again. In this paper, a novel algorithm using a variable payload method is proposed to first identify the compliance coefficients using different payloads and end effector position information. Second, the kinematic parameter errors and link parameters are identified with the given compliance coefficients. Then, the algorithm generates a modified trajectory using the calibrated DH tables for the precision compensation. Simulation and experimental results demonstrate that the positioning accuracy can be improved by 80 % to 90 % even under different payloads. The root mean square, mean, maximum, and standard deviation of the residual errors by using the proposed algorithm could outperform the conventional kinematic algorithm.