Robust centralized fusion Kalman predictors with multiplicative noises, uncertain noise variances, and missing measurements

Abstract

In this paper, the robust centralized fusion Kalman prediction problem is considered for linear discrete-time multisensor systems with multiplicative noises, uncertain noise variances and missing measurements. By introducing two fictitious noises, the system is converted into one with only uncertain noise variances. According to the minimax robust estimation principle, based on the worst-case centralized fusion system with the conservative upper bounds of the noise variances, we present robust centralized fusion time-varying Kalman predictor, and guarantee its actual prediction error variance to have the corresponding minimal upper bounds for all admissible noise variance uncertainties. By use of the Lyapunov equation approach, we prove its robustness. The accuracy relation among the robust local and centralized fusion Kalman predictor is given. Simulation results show the effectiveness and correctness of the proposed results.