Calibration of external linear robot axes using spline interpolation

This paper introduces a method to enhance the accuracy of any robot system that combines an external linear axis with any industrial robot. The developed technique used measured positions and orientations (poses) of the linear axis to identify the existing deviations from an ideal linear axis model in six degrees of freedom (DOF). Based on the measured data B-Splines were used to model the external axis in all six DOF along its entire travel distance. The resulting model allowed the prediction of the linear axis pose with the help of the calculated splines. With this predictable pose the absolute accuracy of any standard industrial robot can be enhanced that is located on the linear axis. For adaptive processes this routine forms the basis of positioning a standard industrial robot regarding its absolute accuracy in combination with the advantageous workspace capability of an external axis. The described method was evaluated and tested on a standard industrial robot system using a standard external linear axis. With this modelling the absolute accuracy of the linear axis was improved close to its repeatability error. All measurements to determine the position and orientation of the linear axis were done with a laser tracker.