Reduced-order estimation Part 2. Smoothing
continuous-timemean- square smoothing errorreduced-order filtersreduced-order fixed-interval smoothingsuboptimality
Continuous-time Markov processes on general state spaces (60J25) Inference from stochastic processes and prediction (62M20) Linear systems in control theory (93C05) Estimation and detection in stochastic control theory (93E10) Filtering in stochastic control theory (93E11) Data smoothing in stochastic control theory (93E14)
[For part I see the paper reviewed above.] This paper is on a new design technique for reduced-order fixed-interval smoothing. The method utilizes the outputs of two reduced-order filters which are combined in an optimal way to minimize the mean-square smoothing error. The full-order case treated using this smoother is shown to result in the well-known Mayne-Fraser smoother. A simple example involving a double-integrator is included to demonstrate the computational saving and the resulting degree of performance suboptimality associated with the use of this technique.
This page was built for publication: Reduced-order estimation Part 2. Smoothing
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3026882)