Entropy-based Consistency Monitoring for Stochastic Integration Filter
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F18%3A43952504" target="_blank" >RIV/49777513:23520/18:43952504 - isvavai.cz</a>
Result on the web
<a href="https://dx.doi.org/10.23919/ICIF.2018.8455462" target="_blank" >https://dx.doi.org/10.23919/ICIF.2018.8455462</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.23919/ICIF.2018.8455462" target="_blank" >10.23919/ICIF.2018.8455462</a>
Alternative languages
Result language
angličtina
Original language name
Entropy-based Consistency Monitoring for Stochastic Integration Filter
Original language description
The paper deals with state estimation of nonlinear stochastic dynamic discrete-time systems with a special focus on the stochastic integration filter. The filter is an instance of Gaussian filters, which for strongly nonlinear systems may provide inconsistent estimates. Primarily, optimistic inconsistent estimates, which overrate quality of the point estimate, are inappropriate in many applications where estimate integrity is crucial. In this paper, a technique for an estimate consistency monitoring for detection of optimistic estimates is proposed based on entropy. For the purpose of the entropy computation, a probabilistic analysis of the stochastic integration filter behavior is carried out. The proposed consistency monitoring is illustrated in a numerical example.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
20205 - Automation and control systems
Result continuities
Project
—
Continuities
N - Vyzkumna aktivita podporovana z neverejnych zdroju
Others
Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of the 21st International Conference on Information Fusion (FUSION 2018)
ISBN
978-0-9964527-6-2
ISSN
—
e-ISSN
neuvedeno
Number of pages
8
Pages from-to
1676-1683
Publisher name
IEEE
Place of publication
Cambridge, UK
Event location
Cambridge, UK
Event date
Jul 10, 2018
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—