Networked Fusion Kalman Filtering With Multiple Uncertainties

This paper investigates the problem of fusion filtering for a class of networked multisensor fusion systems with multiple uncertainties, including sensor failures, stochastic parameter uncertainties, random observation delays, and packet dropouts. A novel model is proposed to describe the random observation delays and packet dropouts, and a robust optimal fusion filter for the addressed networked multisensor fusion systems is designed using the innovation analysis method. The dimension of the designed filter is the same as that of the original system, which helps to reduce computation cost compared with the augmentation method. Moreover, robust reduced-dimension observation-fusion Kalman filters are proposed to further reduce the computation burden. Note that the designed fusion filter gain matrices can be computed off-line, as they depend only on the upper bounds of random delays and on the occurrence probabilities of delays and sensor failures. Some sufficient conditions are presented for stability and optimality of the designed fusion filters, and a steady-state fusion filter is also given for the networked multisensor fusion systems. Simulations show the effectiveness of the proposed fusion filters.