Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks

Yoshinobu Kawahara, Takehisa Yairi, Kazuo Machida

Research output: Contribution to conferencePaper

1 Citation (Scopus)

Abstract

The development of sophisticated anomaly detection and diagnosis methods for spacecraft is one of the important problems in space system operation. Many researches have been carried on this problem, however these depend only on specific resource; knowledge or data. In this paper, we propose a diagnosis method for spacecraft using Dynamic Bayesian Networks which uses both prior- knowledge and data in a natural way in model-acquisition and perform adaptable and in-depth diagnosis by probabilistic reasoning. The proposed method was applied to the artificial telemetry data that simulates the malfunction of thrusters in rendezvous maneuver of spacecraft, and the effectiveness of the method was confirmed.

Original languageEnglish
Pages5823-5830
Number of pages8
Publication statusPublished - Dec 1 2005
Externally publishedYes
EventInternational Astronautical Federation - 56th International Astronautical Congress 2005 - Fukuoka, Japan
Duration: Oct 17 2005Oct 21 2005

Other

OtherInternational Astronautical Federation - 56th International Astronautical Congress 2005
CountryJapan
CityFukuoka
Period10/17/0510/21/05

Fingerprint

Bayesian networks
learning
Failure analysis
Spacecraft
spacecraft
malfunctions
rendezvous
telemetry
maneuvers
Telemetering
acquisition
resources
anomalies
anomaly
method
resource

All Science Journal Classification (ASJC) codes

  • Space and Planetary Science
  • Aerospace Engineering

Cite this

Kawahara, Y., Yairi, T., & Machida, K. (2005). Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks. 5823-5830. Paper presented at International Astronautical Federation - 56th International Astronautical Congress 2005, Fukuoka, Japan.

Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks. / Kawahara, Yoshinobu; Yairi, Takehisa; Machida, Kazuo.

2005. 5823-5830 Paper presented at International Astronautical Federation - 56th International Astronautical Congress 2005, Fukuoka, Japan.

Research output: Contribution to conferencePaper

Kawahara, Y, Yairi, T & Machida, K 2005, 'Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks', Paper presented at International Astronautical Federation - 56th International Astronautical Congress 2005, Fukuoka, Japan, 10/17/05 - 10/21/05 pp. 5823-5830.
Kawahara Y, Yairi T, Machida K. Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks. 2005. Paper presented at International Astronautical Federation - 56th International Astronautical Congress 2005, Fukuoka, Japan.
Kawahara, Yoshinobu ; Yairi, Takehisa ; Machida, Kazuo. / Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks. Paper presented at International Astronautical Federation - 56th International Astronautical Congress 2005, Fukuoka, Japan.8 p.
@conference{7c8e3280aac040b68cfe7ed9cdcc10ae,
title = "Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks",
abstract = "The development of sophisticated anomaly detection and diagnosis methods for spacecraft is one of the important problems in space system operation. Many researches have been carried on this problem, however these depend only on specific resource; knowledge or data. In this paper, we propose a diagnosis method for spacecraft using Dynamic Bayesian Networks which uses both prior- knowledge and data in a natural way in model-acquisition and perform adaptable and in-depth diagnosis by probabilistic reasoning. The proposed method was applied to the artificial telemetry data that simulates the malfunction of thrusters in rendezvous maneuver of spacecraft, and the effectiveness of the method was confirmed.",
author = "Yoshinobu Kawahara and Takehisa Yairi and Kazuo Machida",
year = "2005",
month = "12",
day = "1",
language = "English",
pages = "5823--5830",
note = "International Astronautical Federation - 56th International Astronautical Congress 2005 ; Conference date: 17-10-2005 Through 21-10-2005",

}

TY - CONF

T1 - Fault diagnosis for spacecraft using probabilistic reasoning and statistical learning with dynamic bayesian networks

AU - Kawahara, Yoshinobu

AU - Yairi, Takehisa

AU - Machida, Kazuo

PY - 2005/12/1

Y1 - 2005/12/1

N2 - The development of sophisticated anomaly detection and diagnosis methods for spacecraft is one of the important problems in space system operation. Many researches have been carried on this problem, however these depend only on specific resource; knowledge or data. In this paper, we propose a diagnosis method for spacecraft using Dynamic Bayesian Networks which uses both prior- knowledge and data in a natural way in model-acquisition and perform adaptable and in-depth diagnosis by probabilistic reasoning. The proposed method was applied to the artificial telemetry data that simulates the malfunction of thrusters in rendezvous maneuver of spacecraft, and the effectiveness of the method was confirmed.

AB - The development of sophisticated anomaly detection and diagnosis methods for spacecraft is one of the important problems in space system operation. Many researches have been carried on this problem, however these depend only on specific resource; knowledge or data. In this paper, we propose a diagnosis method for spacecraft using Dynamic Bayesian Networks which uses both prior- knowledge and data in a natural way in model-acquisition and perform adaptable and in-depth diagnosis by probabilistic reasoning. The proposed method was applied to the artificial telemetry data that simulates the malfunction of thrusters in rendezvous maneuver of spacecraft, and the effectiveness of the method was confirmed.

UR - http://www.scopus.com/inward/record.url?scp=34250811410&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34250811410&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:34250811410

SP - 5823

EP - 5830

ER -