Filtragem com previsão-correção para sistemas lineares com saltos Makovianos a tempo discreto

Abstract

In this paper we consider a finite horizon filtering problem for discrete-time Markov jump linear systems. We assume that only the output and the jump parameters are available to the filter and the goal is to obtain a “prediction-correction” formula which is optimal among the Markovian filters. As in the case with no jumps, we show that an optimal filter can be obtained from a set of coupled Riccati difference equations associated to the filtering problem. When there is only one mode of operation our results coincide with the traditional Kalman filter for discrete-time linear systems.