Multi-model information fusion Kalman smoother for time-varying systems

For the multisensor linear discrete time-varying stochastic control systems with multi-model (different local models), three optimal weighted fusion Kalman smoothers weighted by matrices, diagonal matrices and scalars are presented in the linear minimum variance sense, respectively. They are locally optimal and are globally suboptimal. The accuracy of the fusers is higher than that of each local Kalman smoothers, and is lower than that of the centralized fuser. In order to compute the optimal weights, the formula of computing the cross-covariances among local smoothing errors is given. The corresponding steady-state fusion Kalman fusers are also given, which can reduce the on-line computational burden. They can handle the multisensor systems with colored measurement noises. Two Monte Carlo simulation examples for the tracking systems show their effectiveness.