Optimal Observation Of The Multi-sensor Fusion State Estimators And Its Application

For the systems with multisensor, applying the information filtering principle, based on the modified Riccati equation , inverse prediction error variance matrix equation, and inverse filtering error variance matrix equation, the corresponding three fast algorithms for time-varying and steady state Kalman filters based on the centralized measurement fusion are presented , which avoid the high dimensional inverse matrix,and obviously reduce the computational burden. For the two weighted measurement fusion algorithms of multisensor measurement fusion, using the Kalman method, it is proved that compared with the centralized measurement fusion algorithm, they have the global optimality and completely functional equivalence. Based on this, two weighting measurement fusion Kalman estimators(filter, predictor, smoother) and two weighting measurement fusion Wiener state estimators are presented respectively. Compared with the centralized measurement fusion method, by the weighted measurement fusion method, not only the globally optimal estimation can be obtained, but also the computational burden can obviously be reduced, and it is suitable for real time applications. Many Monte Carlo simulation examples show their effectiveness.