Base frame calibration for multi-robot cooperative grinding station by binocular vision

Abstract

Multi-robot coordination grinding need repetitive calibration of the base frame system, and the robot high-density distribution has brought a certain degree of difficulty for the calibration. This paper proposed a new calibration method based on binocular vision by studying existing non-contact calibration method. Firstly, the calibration of the internal and external parameters of the binocular vision system is carried out, then observing space trajectory of robot end-effectors with binocular vision system. Finally, combined with the coordinate transformation theory, the spatial pose relationship between the two robots' base frame is obtained by the designed algorithm. The spatial pose transformation matrix between two ABB grinding robots of our laboratory was determined with this method. Reverse validation experiment's results show that the accuracy of the calibration can reach 0.12mm in three planes of the coordinate axis, and the calibration result is better than other calibration methods.