Statistical Characteriztion of Position Errors of an Ensemble of Robots and Its Applications

Abstract

For the accurate control and implementation of a robot in an integrated manufacturing environment using off-line programming, a knowledge of the absolute positioning accuracy of the robot becomes important. This paper presents a framework which can be used to statistically represent the absolute positioning accuracy for a family of robots. Statistical error measure indices are proposed to represent the position error field over the working space for a family of robots. This error field provides statistical information for the position errors of the end-effector and can be a guide for the determination of the optimal design tolerances of the parts composing of a robot. The second objective of the paper is to introduce a simple interpolation scheme to improve the local position accuracy by teaching one or more task reference frames with which goal positions are mathematically expressed. It will be shown how the method shifts or alters the position error field in order to maintain the desired position accuracy within a desired working area.