A framework for state-space estimation with uncertain models. (English) Zbl 1009.93075

The paper develops a framework for state-space estimation when the parameters of the underlying linear model are subject to uncertainties. The proposed filters are designed to minimize the worst possible regularized residual norm over the class of admissible uncertainties. The author focuses on the following uncertain state-space model \[ x_{i+1}= (F_i+\delta F_i)x_i+ G_i u_i,\quad y_i= H_i x_i+ v_i, \] which is often studied in the literature on robust filtering. Simulation results and comparisons with existing robust filters are provided.


93E10 Estimation and detection in stochastic control theory
93E11 Filtering in stochastic control theory
Full Text: DOI Link