Proceedings of the
9th International Conference of Asian Society for Precision Engineering and Nanotechnology (ASPEN2022)
15 – 18 November 2022, Singapore

Estimation of Kinematic Parameter Errors of 6-axis Serial Robot through Circular Test

Heung-ki Jeon1, Sung-hwan Kweon2, Ki-bum Jung1, Kwang-il Lee3 and Seung-han Yang1,#

1School of Mechanical Engineering, Kyungpook National University, 80, Daehak-ro, Bukgu, Daegu, 41566, South Korea

2Institute of Mechanical Engineering Technology, Kyungpook National University, 80, Daehak-ro, Bukgu, Daegu, 41566, South Korea

3School of Mechanical Automotive Engineering, Kyungil University, 50, Gamasil-gil, Hayang-eup, Gyeongsan, 38428, South Korea


The accuracy of an end-effector pose of a robot is affected by various sources of error, and they can yield improper motion in case of gripping, machining, and so on. There are lots of studies dealing with calibrating the robot in the way of updating kinematic parameters through measuring the end-effector's pose using laser tracker, dial indicator, photogrammetry, and so on. A circular test is one of the fast, simple, and affordable methods for checking a machine's performance, by measuring radial deviation while moving the tool in a circular way. This work aims to check the viability of estimating kinematic parameter errors of the robot from a circular test via by means of simulation. KUKA KR60HA is chosen as a robot, and the kinematic model is established using DH (Denavit-Hartenberg) notations. The error of the end-effector's position is induced by providing randomly-selected, and normally-distributed values into each kinematic parameter. For the test, measurement points on the pre-determined circular path and corresponding joint angles are calculated through inverse kinematics. At each point, poses of the robot are selected in order to make maximum variation in each joint angles against the prior pose. To identify the value of each kinematic parameter, the mathematical relation between the measured radial deviation and kinematic parameter errors is derived. Finally, the estimated kinematic parameters are compared with the designed one to verify the proposed algorithm. Consequently, this work contributes to show feasibility of applying circular test to a 6-axis serial robot for evaluating performance in accuracy.

Keywords: Calibration, Circular test, Kinematic parameter, Serial robot

PDF Download