New method and portable measurement device for the calibration of industrial robots

Abstract

Throughout time, with robotic systems having a more significant impact each day in industrial processes, demand for rapid, repeatable and highly accurate robotic arms has increased, outgrowing the availability of robot calibration options. Even though research on robot calibration has been going on for a long time, implementing a fast, fully automated and affordable calibration system can be hard to find, considering the challenging requirements that a system could bring along with it. This thesis presents an automated calibration method for industrial robots. It is based on the use of a novel, low-cost, wireless, 3D measuring device mounted on the robot end-effector and a portable 3D ball artifact fixed with respect to the robot base. The new device, called TriCal, is essentially a fixture holding three digital indicators (plunger style) with an accuracy of 1.8 μm, the axes of which are orthogonal and intersect at one point, considered to be the robot tool center point (TCP). The artifact contains four 1-inch datum balls, each mounted on heavy-duty riser blocks through stems, and have precisely known relative positions measured on a CMM. The design of the artifact is decided through practical considerations and simulation tests applied through an offline simulation and robot programming software. The measurement procedure contains a physical constraining technique, where the TriCal is probing the datum balls. The method and procedure presented is fully automated and consists of the robot moving its end-effector in such as a way as to perfectly align its TCP with the center of each of the four datum balls, with multiple endeffector orientations. The calibration method and hardware are tested on a six-axis industrial robot (KUKA KR6 R700 sixx). The Modified Denavit-Hartenberg method is used for modeling purposes and the calibration model includes all kinematic and joint stiffness parameters along with an additional parameter defining the relation between two parallel axes. These parameters are later identified using the least-squares method. The joint configurations for calibration are decided through an observability index calculation from an initially generated configuration pool. The efficiency of the new calibration system is validated by measuring the accuracy of the robot after calibration in 500 nearly random end-effector poses using a laser tracker. The same validation is performed after the robot was calibrated using measurements from the laser tracker only. Results show that both measurement methods lead to similar accuracy improvements, with the TriCal yielding maximum position errors of 0.624 mm, and mean position errors of 0.326 mm.

Date	18 Nov 2020
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Ilian Bonev (Supervisor)