The problem of fusion of local estimates is considered. An optimal mean-square linear combination (fusion formula) of an arbitrary number of local vector estimates is derived. The derived result holds for all dynamic systems with measurements. In particular, for scalar uncorrelated local estimates, the fusion formula represents the well-known result in statistics. The fusion formula is applied to fusion of local Kalman estimates in multisensor filtering problem. Examples demonstrate high accuracy of the proposed fusion formula. © 2007 Elsevier B.V. All rights reserved.
Shin, V., Shevlyakov, G., & Kim, K. (2007). A new fusion formula and its application to continuous-time linear systems with multisensor environment. Computational Statistics and Data Analysis, 52(2), 840–854. https://doi.org/10.1016/j.csda.2006.11.031