Model-based fault detection for the DELFI-N3XT attitude determination system

Napoleon E. Cornejo; Rouzbeh Amini; Georgi Gaydadjiev

doi:10.1109/AERO.2010.5446801

Model-based fault detection for the DELFI-N3XT attitude determination system

Napoleon E. Cornejo, Rouzbeh Amini, Georgi Gaydadjiev

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

10 Citations (Scopus)

Abstract

The Delfi-n3Xt nanosatellite is the second Dutch university satellite currently being developed at the Delft University of Technology. In its design, the Attitude Determination System (ADS) will be pivotal for optimal power point tracking to adequately provide the energy needed for normal operation and charging of the batteries. In this paper we explore a fault detection mechanism for the ADS based on the Unscented Kalman Filter (UKF) state estimator which has been successfully integrated into the simulation and modelling environment. The UKF provides a more computationally efficient estimator than traditional Kalman Filter variants. Faults introduced in the system include changes in the noise model and stuck-at-0 faults, resulting in disturbances in the output of the filter. Parameters of the filter are varied and the behaviour of the outcoming residuals is analyzed to evaluate its effectiveness in the detection of these errors.

Original language	English
Title of host publication	2010 IEEE Aerospace Conference Proceedings
Publisher	IEEE
Pages	1-8
Number of pages	8
ISBN (Print)	978-1-4244-3887-7
DOIs	https://doi.org/10.1109/AERO.2010.5446801
Publication status	Published - 2010
Event	2010 IEEE Aerospace Conference - Big Sky, MT, United States Duration: 6 Mar 2010 → 13 Mar 2010

Conference

Conference	2010 IEEE Aerospace Conference
Country/Territory	United States
City	Big Sky, MT
Period	6/03/10 → 13/03/10

Access to Document

10.1109/AERO.2010.5446801

Cite this

@inproceedings{3d7b51bc912c4df09618c13f2dde2cf9,

title = "Model-based fault detection for the DELFI-N3XT attitude determination system",

abstract = "The Delfi-n3Xt nanosatellite is the second Dutch university satellite currently being developed at the Delft University of Technology. In its design, the Attitude Determination System (ADS) will be pivotal for optimal power point tracking to adequately provide the energy needed for normal operation and charging of the batteries. In this paper we explore a fault detection mechanism for the ADS based on the Unscented Kalman Filter (UKF) state estimator which has been successfully integrated into the simulation and modelling environment. The UKF provides a more computationally efficient estimator than traditional Kalman Filter variants. Faults introduced in the system include changes in the noise model and stuck-at-0 faults, resulting in disturbances in the output of the filter. Parameters of the filter are varied and the behaviour of the outcoming residuals is analyzed to evaluate its effectiveness in the detection of these errors.",

author = "Cornejo, {Napoleon E.} and Rouzbeh Amini and Georgi Gaydadjiev",

year = "2010",

doi = "10.1109/AERO.2010.5446801",

language = "English",

isbn = "978-1-4244-3887-7",

pages = "1--8",

booktitle = "2010 IEEE Aerospace Conference Proceedings",

publisher = "IEEE",

address = "United States",

note = "2010 IEEE Aerospace Conference ; Conference date: 06-03-2010 Through 13-03-2010",

}

TY - GEN

T1 - Model-based fault detection for the DELFI-N3XT attitude determination system

AU - Cornejo, Napoleon E.

AU - Amini, Rouzbeh

AU - Gaydadjiev, Georgi

PY - 2010

Y1 - 2010

N2 - The Delfi-n3Xt nanosatellite is the second Dutch university satellite currently being developed at the Delft University of Technology. In its design, the Attitude Determination System (ADS) will be pivotal for optimal power point tracking to adequately provide the energy needed for normal operation and charging of the batteries. In this paper we explore a fault detection mechanism for the ADS based on the Unscented Kalman Filter (UKF) state estimator which has been successfully integrated into the simulation and modelling environment. The UKF provides a more computationally efficient estimator than traditional Kalman Filter variants. Faults introduced in the system include changes in the noise model and stuck-at-0 faults, resulting in disturbances in the output of the filter. Parameters of the filter are varied and the behaviour of the outcoming residuals is analyzed to evaluate its effectiveness in the detection of these errors.

AB - The Delfi-n3Xt nanosatellite is the second Dutch university satellite currently being developed at the Delft University of Technology. In its design, the Attitude Determination System (ADS) will be pivotal for optimal power point tracking to adequately provide the energy needed for normal operation and charging of the batteries. In this paper we explore a fault detection mechanism for the ADS based on the Unscented Kalman Filter (UKF) state estimator which has been successfully integrated into the simulation and modelling environment. The UKF provides a more computationally efficient estimator than traditional Kalman Filter variants. Faults introduced in the system include changes in the noise model and stuck-at-0 faults, resulting in disturbances in the output of the filter. Parameters of the filter are varied and the behaviour of the outcoming residuals is analyzed to evaluate its effectiveness in the detection of these errors.

UR - http://www.scopus.com/inward/record.url?scp=77952864513&partnerID=8YFLogxK

U2 - 10.1109/AERO.2010.5446801

DO - 10.1109/AERO.2010.5446801

M3 - Conference contribution

AN - SCOPUS:77952864513

SN - 978-1-4244-3887-7

SP - 1

EP - 8

BT - 2010 IEEE Aerospace Conference Proceedings

PB - IEEE

T2 - 2010 IEEE Aerospace Conference

Y2 - 6 March 2010 through 13 March 2010

ER -

Model-based fault detection for the DELFI-N3XT attitude determination system

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this