Filter, Multi-sensor Information Fusion Based On Kalman Filtering Method

For multisensor systems with different local dynamic models and with correlated noises, using Kalman filtering method, based on the Riccati equation, under the linear minimum variance optimal information fusion criterion, the distributed information fusion steady-state Kalman filter, predictor, smoother and the corresponding Wiener state estimators weighted by matrices, diagonal matrices ,and scalars, are presented, respectively . The Lyapunov equations of computing the cross-covariances among local estimation errors are presented, which can be applied to compute the optimal weights. Based on this, by the augmented state method, the multisensor distributed information fusion Wiener signal estimators and Wiener signal deconvolution estimators are also presented. The simulation example in tracking systems and the numerical simulation examples show their effectiveness, and show that the accuracy distinction of three weighted fusion algorithms is not obvious. Therefore, employing the fusion algorithm weighted by scalars can obviously reduce the computational burden, and is suitable for real time applications.