Accurate Density-Weighted Convolution for Point-Mass Filter and Predictor

dc.contributor.authorDuník, Jindřich
dc.contributor.authorStraka, Ondřej
dc.contributor.authorMatoušek, Jakub
dc.contributor.authorBrandner, Marek
dc.date.accessioned2022-02-28T11:00:21Z
dc.date.available2022-02-28T11:00:21Z
dc.date.issued2021
dc.description.abstract-translatedThis paper deals with the Bayesian state estimation of nonlinear stochastic dynamic systems. The stress is laid on the numerical solution to the Chapman-Kolmogorov equation, which governs the prediction step of the point-mass filter and predictor, using the convolution. A novel density-weighted convolution is proposed, which provides an accurate predictive probability density function even for models with small state noise, where the standard solution fails. Two implementations of the solution are proposed, theoretically analyzed, and evaluated in a numerical study.en
dc.format11 s.cs
dc.format.mimetypeapplication/pdf
dc.identifier.citationDUNÍK, J. STRAKA, O. MATOUŠEK, J. BRANDNER, M. Accurate Density-Weighted Convolution for Point-Mass Filter and Predictor. IEEE Transactions on Aerospace and Electronic Systems, 2021, roč. 57, č. 6, s. 3574-3584. ISSN: 0018-9251cs
dc.identifier.document-number725819700005
dc.identifier.doi10.1109/TAES.2021.3079568
dc.identifier.issn0018-9251
dc.identifier.obd43933468
dc.identifier.uri2-s2.0-85105868526
dc.identifier.urihttp://hdl.handle.net/11025/47006
dc.language.isoenen
dc.project.IDSGS-2019-020/Rozvoj a využití kybernetických systémů identifikace, diagnostiky a řízení 4cs
dc.publisherIEEEen
dc.relation.ispartofseriesIEEE Transactions on Aerospace and Electronic Systemsen
dc.rightsPlný text je přístupný v rámci univerzity přihlášeným uživatelům.cs
dc.rights© IEEEen
dc.rights.accessrestrictedAccessen
dc.subject.translatedstate estimationen
dc.subject.translatedBayesian inferenceen
dc.subject.translatednonlinear systemsen
dc.subject.translatedpoint-mass filteren
dc.titleAccurate Density-Weighted Convolution for Point-Mass Filter and Predictoren
dc.typečlánekcs
dc.typearticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen

Files