Scalar-weighted fusion estimators for systems with multiple sensors and multiple delayed measurements

This paper is concerned with the distributed fusion estimation problem for discrete-time stochastic linear system with multiple sensors having multiple delayed measurements and correlated noise. Distributed weighted fusion optimal estimators are given based on local optimal estimators from single sensor and the optimal scalar-weighted fusion algorithm in the linear minimum variance sense. Compared with the augmented optimal estimators, the distributed fusion estimators with scalar weights are more reliable and have the reduced computation cost since they have twith multiple sensors having multiple delayed measurements and correlated noise. Distributed weighted fusion optimal estimators are given based on local optimal estimators from single sensor and the optimal scalar-weighted fusion algorithm in the linear minimum variance sense. Compared with the augmented optimal estimators, the distributed fusion estimators with scalar weights are more reliable and have the reduced computation cost since they have the parallel structure. The estimation error crosshe parallel structure. The estimation error cross-covariance matrices between any two-sensor subsystems are derived. Applying to a tracking system with three sensors shows the effectiveness.