Online robot calibration based on hybrid sensors using Kalman Filters

Abstract

This paper presents an online robot self-calibration method based on an inertial measurement unit (IMU) and a position sensor. In this method, a position marker and an IMU are required to rigidly attach to the robot tool, which makes it possible to obtain the position of the manipulator from the position sensor and the orientation from the IMU in real time. An efficient approach which incorporates Kalman Filters (KFs) to estimate the position and the orientation of the manipulator is proposed in this paper. Using these pose (orientation and position) estimation methods will result in improving the reliability and accuracy of pose measurements. Finally, an Extended Kalman Filter (EKF) is used to estimate kinematic parameter errors. Compared with the existing self-calibration methods, the greatest advantage of this method is that it does not need any complex steps, such as camera calibration, corner detection and laser tracking, which makes the proposed robot calibration procedure more autonomy in a dynamic manufacturing environment. What׳s more, reduction in complex steps leads to improving the accuracy of calibration. Experimental studies on a GOOGOL GRB3016 robot show that this proposed method has better accuracy, convenience and effectiveness.