Efficient Implementation of Marginal Particle Filter by Functional Density Decomposition

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F22%3A43966079" target="_blank" >RIV/49777513:23520/22:43966079 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.23919/FUSION49751.2022.9841367" target="_blank" >https://doi.org/10.23919/FUSION49751.2022.9841367</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.23919/FUSION49751.2022.9841367" target="_blank" >10.23919/FUSION49751.2022.9841367</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Efficient Implementation of Marginal Particle Filter by Functional Density Decomposition

Popis výsledku v původním jazyce

The paper considers the solution to the state estimation problem of nonlinear dynamic stochastic systems by the particle filters. It focuses on the marginal particle filter algorithms which generate samples directly in the marginal space for the recent state. Their standard implementation calculates the sample weights by combining the samples from two consecutive time instants in the transition and proposal density function evaluations. This results in computational complexity quadratic in sample size. The paper proposes an efficient implementation of the marginal particle filter for which a functional tensor decomposition of the transition and proposal densities is calculated. The computational complexity of the proposed implementation is linear in sample size and the decomposition rank can be used to achieve a trade-off between accuracy and computational costs. The balance between the complexity and the estimate quality can be tuned by selecting the rank of the decomposition. The proposed implementation is demonstrated using two numerical examples with a univariate non-stationary growth model and terrain-aided navigation scenario.

Název v anglickém jazyce

Efficient Implementation of Marginal Particle Filter by Functional Density Decomposition

Popis výsledku anglicky

The paper considers the solution to the state estimation problem of nonlinear dynamic stochastic systems by the particle filters. It focuses on the marginal particle filter algorithms which generate samples directly in the marginal space for the recent state. Their standard implementation calculates the sample weights by combining the samples from two consecutive time instants in the transition and proposal density function evaluations. This results in computational complexity quadratic in sample size. The paper proposes an efficient implementation of the marginal particle filter for which a functional tensor decomposition of the transition and proposal densities is calculated. The computational complexity of the proposed implementation is linear in sample size and the decomposition rank can be used to achieve a trade-off between accuracy and computational costs. The balance between the complexity and the estimate quality can be tuned by selecting the rank of the decomposition. The proposed implementation is demonstrated using two numerical examples with a univariate non-stationary growth model and terrain-aided navigation scenario.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

<a href="/cs/project/GA22-11101S" target="_blank" >GA22-11101S: Tenzorový rozklad v aktivní diagnostice poruch pro stochastické rozlehlé systémy</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 25th International Conference on Information Fusion, FUSION 2022

ISBN

978-1-73774-972-1

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

1-8

Název nakladatele

IEEE

Místo vydání

Linköping, Sweden

Místo konání akce

Linköping, Sweden

Datum konání akce

4. 7. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000855689000167