Robust Parametrized Minimum-Variance Filtering for Uncertain Systems with Unknown Inputs

This paper considers the minimum-variance estimation for uncertain systems with unknown inputs that affect both the system model and the measurements. By making use of a constrained optimization method, a parametrized filter structure, and the enforcement of a minimum state- error variance property, a robust parametrized minimum- variance filter is derived for uncertain systems to achieve an optimal compromise between a robust version of the optimal unbiased minimum-variance filter and a robust Kalman filter. A numerical example is included in order to illustrate the usefulness of the proposed results.