An enhanced kinematic model for calibration of robotic machining systems with parallelogram mechanisms

Abstract Kinematic calibration presents an efficient way to improve the absolute accuracy of robotic machining systems with parallelogram mechanisms. However, the mixed chain structure, i.e., the open loop of the industrial robot and the closed loop of the kinematic parallelogram, complicates the identification of the geometrical parameters. By segmentation of the closed loop, the mixed chain structure could be reconstructed into two individually open chains, i.e., one short chain and one long chain, for modeling of kinematics. This paper proposes an improved kinematic calibration method based on the short chain to identify the geometrical parameter errors in the open and closed loops and evaluates the efficiency of the kinematic calibration models based on the two open chains. Firstly, the error model in the closed loop is derived based on the direct linearization of the closed-loop vector equations, and then is integrated into the short chain model to establish a global error model. Subsequently, the redundant parameters in the model are easily removed. The comparison between the proposed calibration model and the model based on the long chain is conducted under a calibration experiment for a robotic machining system. The results show that the proposed calibration model is more computationally efficient with almost the same accuracy, which is better for kinematic calibration of the practical system. Finally, a compensation experiment with a particular machining path is performed to verify the effectiveness of the proposed calibration model.