Work

Unscented Kalman Observer

Year: 2019

Type: conference-paper

Abstract: The extended Kalman filter is an exponentially converging observer as soon as it is written in a canonical form of observability and in its high-gain form. It is shown that unlike extended Kalman filter, unscented Kalman filter can not be an exponentially converging observer. We propose a slight modification of the unscented Kalman filter to build an exponentially converging observer called unscented Kalman observer. Performances of this new observer are illustrated on an example of geolocation problem.

Source: International Conference on Systems and Control (ICSC)

DOI: 10.1109/icsc47195.2019.8950505

Authors: Assia Daid, Eric Busvelle, Mohamed Aidene

Institutions: Université de Toulon, Laboratoire d’Informatique et Systèmes, Mouloud Mammeri University of Tizi-Ouzou

Language: English

Topic: Target Tracking and Data Fusion in Sensor Networks

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W3000717434

FWCI: 0.326

Cites: 27

Cited by: 4

Related to: 10