Residual index for measurement configuration optimization in robot kinematic calibration

To improve the efficiency and accuracy of kinematic calibration, the selection of measurement configurations is an important issue. In previous research, optimal measurement configurations mainly are selected by maximizing observability indices. However, the traditional observability indices only focus on the identification efficiency of the error parameters, while the purpose of robot kinematic calibration is to improve accuracy. To solve the inconsistency of the purpose between the observability index and calibration, the concept of the residual index to represent the residual distribution of the end effector after robot kinematic calibration with the measurement noise is proposed. Based on the quadratic form minimization of residuals, this article defines a specific residual index, Or, which is dimensionless and strictly better with the increase of measurement configurations. The indices are used to select measurement configurations in the kinematic calibration of a 5-DOF 2UPU/SP-RR hybrid robot, and the calibration results show that the proposed residual index is better than the traditional indices in the accuracy and stability of the end effector residual.