Varying Mass Estimation and Force Ripple Compensation using Extended Kalman Filter for Linear Motor Systems

Abstract

In many industrial fields, the mass information of a moving system is important and necessary to prevent undesired motion or failure and to control the system in its desired trajectory. One simple solution could be direct measurement of the mass using a sensor such as force sensor and accelerometer. However, it requires additional cost increase. In addition, it is not easy to measure the mass of a moving part in many cases. For those reasons, in this research, an online varying mass estimation algorithm is designed using an Extended Kalman Filter (EKF) without any additional sensors. Furthermore, the lumped disturbance compensating algorithm, which was designed by the authors in the previous research using EKF, is combined to obtain further position tracking performance. The effectiveness of the suggested method is validated through simulations. Additional verification with experiments is planned for future work.