The Flight Control Team Multi-Agent System (FCTMAS) study, funded by the European Space Agency (ESA), has investigated the use of multiagent systems in supporting flight control teams in routine operations. One of the scientific challenges of the FCTMAS study has been the detection of anomalies relative to a space system only on the basis of identified deviations from the nominal trends of single measurable variables. In this paper, we discuss how we addressed this challenge by looking for the best structure that aggregates a given set of models, each one returning the anomaly probability of a single measurable variable, under the assumption that there is no a priori knowledge about the structure of the space system nor about the relationships between the variables. Experiments are conducted on data of the Cryosat-2 satellite and their results are eventually summarized as a set of guidelines.

Aggregating models for anomaly detection in space systems: Results from the FCTMAS study

Amigoni, Francesco;Ferrari Dacrema, Maurizio;Lavagna, Michèle;
2019-01-01

Abstract

The Flight Control Team Multi-Agent System (FCTMAS) study, funded by the European Space Agency (ESA), has investigated the use of multiagent systems in supporting flight control teams in routine operations. One of the scientific challenges of the FCTMAS study has been the detection of anomalies relative to a space system only on the basis of identified deviations from the nominal trends of single measurable variables. In this paper, we discuss how we addressed this challenge by looking for the best structure that aggregates a given set of models, each one returning the anomaly probability of a single measurable variable, under the assumption that there is no a priori knowledge about the structure of the space system nor about the relationships between the variables. Experiments are conducted on data of the Cryosat-2 satellite and their results are eventually summarized as a set of guidelines.
2019
Intelligent Autonomous Systems 15
978-3-030-01369-1
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1080212
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact