Two-sensor Information Fusion Optimal And Self-correcting Filter

Under the linear minimum variance optimal information fusion criterion, by using the modern time series analysis method, based on the autoregressive moving average (ARMA) innovation model, the optimal and self-toning information fusion Kalman filters weighted by scalars, vectors (diagonal matrices) and matrices are presented for two-sensor systems, respectively, where the correlation between estimating errors is considered, the corresponding weighting formulas of optimal fused estimation are given, and the ARMA. signal optimal and self-tuning information fusion Wiener filters with white observation noises, the optimal and the self-tuning information fusion white noise Wiener deconvolution filters, and the optimal and the self-tuning information fusion Wiener deconvolution filters are also presented. The proposed methods avoide the Riccati equation and Diophantine equation and can reduce the on-line computational burden. Specially, they can be applied to solve the self-tuning information fusion filtering problems for systems with unknown model parameters and noise statistics. Large numbers of simulation examples show their effectiveness.