Student-t process quadratures for filtering of non-linear systems with heavy-tailed noise
(2017) 20th International Conference on Information Fusion, FUSION 2017- Abstract
- The aim of this article is to design a moment transformation for Student-t distributed random variables, which is able to account for the error in the numerically computed mean. We employ Student-t process quadrature, an instance of Bayesian quadrature, which allows us to treat the integral itself as a random variable whose variance provides information about the incurred integration error. Advantage of the Student-t process quadrature over the traditional Gaussian process quadrature, is that the integral variance depends also on the function values, allowing for a more robust modelling of the integration error. The moment transform is applied in nonlinear sigma-point filtering and evaluated on two numerical examples, where it is shown to... (More)
- The aim of this article is to design a moment transformation for Student-t distributed random variables, which is able to account for the error in the numerically computed mean. We employ Student-t process quadrature, an instance of Bayesian quadrature, which allows us to treat the integral itself as a random variable whose variance provides information about the incurred integration error. Advantage of the Student-t process quadrature over the traditional Gaussian process quadrature, is that the integral variance depends also on the function values, allowing for a more robust modelling of the integration error. The moment transform is applied in nonlinear sigma-point filtering and evaluated on two numerical examples, where it is shown to outperform the state-of-the-art moment transforms. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/70c91bc4-4234-4baa-9a9a-5e63a6dec118
- author
- Prüher, Jakob ; Tronarp, Filip LU ; Karvonen, Toni ; Särkkä, Simo and Straka, Ondrej
- publishing date
- 2017
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- host publication
- 20th International Conference on Information Fusion (Fusion)
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 20th International Conference on Information Fusion, FUSION 2017
- conference location
- Xian, China
- conference dates
- 2017-07-10 - 2017-07-13
- external identifiers
-
- scopus:85029415650
- ISBN
- 978-1-5090-4582-2
- 978-0-9964-5270-0
- DOI
- 10.23919/ICIF.2017.8009742
- language
- English
- LU publication?
- no
- id
- 70c91bc4-4234-4baa-9a9a-5e63a6dec118
- date added to LUP
- 2023-08-20 22:42:02
- date last changed
- 2024-08-25 18:56:49
@inproceedings{70c91bc4-4234-4baa-9a9a-5e63a6dec118,
abstract = {{The aim of this article is to design a moment transformation for Student-t distributed random variables, which is able to account for the error in the numerically computed mean. We employ Student-t process quadrature, an instance of Bayesian quadrature, which allows us to treat the integral itself as a random variable whose variance provides information about the incurred integration error. Advantage of the Student-t process quadrature over the traditional Gaussian process quadrature, is that the integral variance depends also on the function values, allowing for a more robust modelling of the integration error. The moment transform is applied in nonlinear sigma-point filtering and evaluated on two numerical examples, where it is shown to outperform the state-of-the-art moment transforms.}},
author = {{Prüher, Jakob and Tronarp, Filip and Karvonen, Toni and Särkkä, Simo and Straka, Ondrej}},
booktitle = {{20th International Conference on Information Fusion (Fusion)}},
isbn = {{978-1-5090-4582-2}},
language = {{eng}},
publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
title = {{Student-t process quadratures for filtering of non-linear systems with heavy-tailed noise}},
url = {{http://dx.doi.org/10.23919/ICIF.2017.8009742}},
doi = {{10.23919/ICIF.2017.8009742}},
year = {{2017}},
}