Simultaneous state and parameter estimation based on AUMV estimator and IBR-KF

Abstract

We proposed a new robust states and parameters estimator for nonlinear systems in this paper. The proposed estimator is derived based on Approximated Unbiased Minimum Variance (AUMV) estimator [3] and Intrinsically Bayesian Robust Kalman filter (IBR-KF) [4]. So, the proposed method takes over the merits of both method. That is, the estimated states are not influenced by the parameter estimation error as in the AUMV estimator, and the estimated parameter is robust against observation noise like the IBR-KF.We confirm the validity of the proposed methods by numerical simulations.