An adaptive risk-sensitive filtering method for Markov jump linear systems with uncertain parameters

In this paper, an adaptive risk-sensitive multiple-model filtering method which relaxes the restrictive assumption that risk-sensitive parameter is chosen as a prior is proposed for a class of discrete-time Markov jump linear systems (MJLSs) with uncertain parameters. Some analysis is presented to illustrate the essential effect of the risk sensitivity added into the filtering process and show the intrinsic reasons for the improvement of robustness. Then, a quite useful principle is developed to obtain the risk-sensitive parameter using the measurements in an online fashion. To avoid overregulation under mismatched modes and mitigate the problem of smearing the feature of each model, a minimization mechanism is resorted to. Computer simulations are presented to reveal the effectiveness of our method.