Proceedings-Artikel
On the Criteria for Configurations Selection in Robot Calibration
Henry Carrillo; Oliver Birbach; Holger Täubig; Berthold Bäuml; Udo Frese; JosĂ© A. Castellanos
In: Proceedings of the IEEE International Conference on Robotics and Automation. IEEE International Conference on Robotics and Automation (ICRA-13), May 6-10, Karlsruhe, Germany, IEEE, 5/2013.
Abstract
This paper studies the criteria for configurations selection in a robot calibration task. We focus only on the robot’s configuration because the calibration is conceived as an automatic and self-contained procedure (i.e. without external tools). Specifically, we experiment with the active calibration of a multi-sensorial humanoid’s upper body and report a predominance of the determinant based criteria when a greedy selection is used. In addition to the criteria comparison, we also propose a new criterion for configuration selection. Its novelty stems from a direct treatment of the robot’s end-effector tool variance, in contrast to previous approaches that target the variance indirectly via the calibration parameters. Our objective function is derived as a compact formulation of the mean error of the robot’s end-effector tool, from which its variance can be computed using traditional criteria stemming from the Theory of Optimum Experimental Designs (e.g. A optimality).