A fast mixed distributed fusion for systems with random parameter matrices and correlated noises

A fast mixed distributed fusion algorithm for the system with random parameter matrices and correlated noises is proposed in this paper. First, It has fewer estimators because every local estimator can handle more than one sensor measurement data and also has robustness and reliability. Then, to reduce the computational complexity significantly, a sequential fast covariance intersection(SFCI) approach is applied to fuse the local filter estimations. It avoids the calculation of the cross-covariance matrices and the optimization of the cost function. Finally, a numerical example is given to verify the usefulness of the proposed filter algorithm.