Low-Complexity Fusion Estimation Algorithms for Multisensor Dynamic Systems

Seokhyoung LEE  Vladimir SHIN  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E92-A   No.11   pp.2910-2916
Publication Date: 2009/11/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E92.A.2910
Print ISSN: 0916-8508
Type of Manuscript: PAPER
Category: Communication Theory and Signals
estimation,  Kalman filtering,  fusion formula,  multisensor,  Cholesky factorization,  computational complexity,  

Full Text: PDF>>
Buy this Article

This paper focuses on fusion estimation algorithms weighted by matrices and scalars, and relationship between them is considered. We present new algorithms that address the computation of matrix weights arising from multidimensional estimation problems. The first algorithm is based on the Cholesky factorization of a cross-covariance block-matrix. This algorithm is equivalent to the standard composite fusion estimation algorithm however it is low-complexity. The second fusion algorithm is based on an approximation scheme which uses special steady-state approximation for local cross-covariances. Such approximation is useful for computing matrix weights in real-time. Subsequent analysis of the proposed fusion algorithms is presented, in which examples demonstrate the low-computational complexity of the new fusion estimation algorithms.