Measurement Fusion Kalman Filters for Descriptor Stochastic Systems

Centralized measurement fusion and weighted measurement fusion are two main methods for multi-sensor data fusion based on Kalman filtering. The measurement fusion state estimation problem was considered for descriptor stochastic systems. Two kinds of multi-sensor measurement fused state Kalman filters were proposed. The effectiveness of the proposed algorithms was demonstrated by numerical examples. And the functional equivalence between two fused methods was verified under the assumption that the sensors for fused data fusion have identical measurement matrices, i.e. the Kalman filters obtained by two methods are numerically equal.