Enhanced robust Kalman predictor for discrete-time systems with uncertain correlated noises

This paper presents an enhanced robust predictor for uncertain discrete-time systems. Besides uncertainties in both state and output matrices, it is also permitted dynamic and measurement noises to be correlated with unknown correlation covariance. All uncertainties in the proposed model are time-varying and supposed norm-bounded. The filter is obtained minimizing an upper bound of the variance error estimation, that is, the design leads to a guaranteed cost for all allowed uncertainties. Simulation examples are provided to show the performance of the enhanced estimator.