Fixed Interval Optimal Estimation for Linear Discrete-Time Markovian Jump Systems by Maximum Likelihood Approach

Abstract

In this paper we study the optimal state estimation problems for a class of linear discrete-time Markovian jump systems. We adopt maximum likelihood (ML) approach and stochastic variational calculus method to derive forms of dynamic estimators on the fixed time interval. The necessary condition for the solvability of the optimal state estimation problems are given by the coupled Riccati difference equations with initial conditions.