Abstract: Objectives Addressing the issue of the attitude adjustment effect of the shafting docking system being affected by multiple error sources, a mapping model between error sources and attitude adjustment errors is established to identify the geometric error parameters of the system Methods Based on the vector method, a spatial closed-loop vector chain for each component of the system is established, and differential processing is applied to obtain the error model. An improved Levenberg-Marquardt algorithm is used to identify geometric error parameters. To verify the accuracy of the model and identification algorithm, the system's geometric error is simulated in a preset actual measurement configuration. The geometric error parameters are identified through the established error model and calibration algorithm, and the identification results are used to correct the kinematic model. Finally, the identification accuracy and the mean value of pose residuals are used as verification indicators.. Results The simulation results show that the average identification accuracy of the error parameters is 90.33%. After model correction, the Y-direction position error is reduced by an average of 80.55%, the Z-direction position error is reduced by an average of 59.7%, and the angle error is reduced by 72.4%. The angle error remains stable..Conclusions The established error model and identification algorithm are accurate and effective.