Robust minimum variance filtering with an H_∞ error bound

This paper studies the robust minimum variance filtering problem for linear continuous-time systems that are subjected to norm-bounded parameter uncertainty and impinged by unmodelled external disturbances. An upper bound for the worst-case H₂ performance index is obtained under the condition of satisfying a prespecified H_∞ estimation error constraint, and the optimal filter is designed in the sense of minimizing the bound. The filter derived is guaranteed to satisfy a prespecified H_∞ error bound for all admissible parameter uncertainties, while providing an optimized bound for the worst-case H₂ performance.