Presentation 2004/10/12
Stydy on Spacecraft Diagnosis Using Dynamic Bayesian Networks
Yoshinobu KAWAHARA, Takehisa YAIRI, Kazuo MACHIDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Realization of sophisticated novelty detection / diagnostic method is one of the important problems in space system research. In this study, we propose new diagnostic method based on filtering using Dynamic Bayesian Networks (DBNs). Using observation data, this diagnostic method makes it possible to detect significant changes from normal state and discriminate its place and causes. Furthermore, this method has an advantage of large range diagnose capavility because of the rich expressiveness of DBNs, and adaptablity that can correct the model using statistical learning so that it may fit the actual action of a spacecraft system. In this paper, its availability is shown by simulation test.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) diagnosis / spacecraft / Dynamic Bayesian Networks (DBNs) / filtering
Paper # NC2004-72
Date of Issue

Conference Information
Committee NC
Conference Date 2004/10/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Stydy on Spacecraft Diagnosis Using Dynamic Bayesian Networks
Sub Title (in English)
Keyword(1) diagnosis
Keyword(2) spacecraft
Keyword(3) Dynamic Bayesian Networks (DBNs)
Keyword(4) filtering
1st Author's Name Yoshinobu KAWAHARA
1st Author's Affiliation Faculty of Engineering, Tokyo University()
2nd Author's Name Takehisa YAIRI
2nd Author's Affiliation Faculty of Engineering, Tokyo University
3rd Author's Name Kazuo MACHIDA
3rd Author's Affiliation Faculty of Engineering, Tokyo University
Date 2004/10/12
Paper # NC2004-72
Volume (vol) vol.104
Number (no) 349
Page pp.pp.-
#Pages 6
Date of Issue