The Robust Finite-Horizon Kalman Filtering for Multiple Sensors with Stochastic Missing Measurements

Abstract

We consider the filtering problem for norm-bounded uncertain discrete dynamic systems with measurements having different stochastic failure rates. For tackling the uncertainties of the covariance matrices of state and its estimation error simultaneously, their upper bounds containing a scaling parameter are derived, and then, a robust finite-horizon filtering minimizing the upper bound of estimation error covariance is proposed. An illustrative example is included to validate the good performance of the proposed robust filtering.