Kalman Filter for Steer-by-Wire Disturbance Estimation

This paper addresses a practical problem in steer-by-wire systems: mitigating high-frequency disturbances caused by driver input. The use of a Kalman filter is a well-established technique for state estimation, and its application to this specific problem is novel. The paper's contribution lies in the design and evaluation of a Kalman filter-based disturbance observer that estimates driver torque using only motor state measurements, avoiding the need for costly torque sensors. The comparison of linear and nonlinear Kalman filter variants and the analysis of their performance in handling frictional nonlinearities are valuable. The simulation-based validation is a limitation, but the paper acknowledges this and suggests future work.