Application of reduced state estimation to multisensor fusion with out-of-sequence measurements

A filtering application of processing multisensor measurements with delays is considered. Because of delays, measurements fed by geographically dispersed sensors to a processing site may arrive out of time sequence. Unlike smoothing or filtering, optimal processing of an out-of-sequence measurement is not a standard problem in filtering theory for which a definitive approach has yet to be developed. An optimal reduced state estimator, derived in previous work, is applied to this problem. A simulation example of multisensor fusion is presented, in which one sensor feeds highly accurate, but delayed, measurements to be fused with a second sensor's less accurate measurements having no delay. We demonstrate a uniform improvement in performance for this algorithm over two traditional approaches.