A new Minimum Variance Observer for Stochastic LPV systems with Unknown Inputs * * This work is supported by the CNES (Centre National d’Etudes spatiales), France

摘要

This paper is dedicated to the design of a state estimator for discrete-time Linear Parameter Varying (LPV) systems affected by unknown inputs and random Gaussian noises. Contrary to the existing work, the observer designed in this paper takes measures at several time steps into account in order to improve the performance (in terms of minimizing the variance estimation error). This approach is based on combining the classical Kalman Filter with the design strategies of deterministic observer for LPV systems in deterministic framework. Then, as an extension of this result, the observer is used for estimation of LPV systems without unknown inputs when state noises have a very high variance in comparison to the measurement noises. Simulation results are presented to illustrate the effectiveness of the proposed approach.