Measurement fusion information filter for multisensor descriptor system with correlated noises

Abstract

For the multisensor linear stochastic descriptor system with correlated noises, the fused measurement can be obtained based on the weighted least square method and the fused descriptor system can be transformed into a new canonical form applying singular value decomposition (SVD) method. Then, the measurement fusion information filter of the original multisensor descriptor system is presented. The obtained measurement fusion information filter has global optimality, and can avoid computing these cross-variances of the local Kalman filters. A simulation example about 3-sensors stochastic descriptor system verifies the effectiveness.