Explainable anomaly detection in spacecraft telemetry

Cuéllar, Sara; Santos, Matilde; Alonso, Fernando; Fabregas, Ernesto; Farias, Gonzalo

doi:10.1016/j.engappai.2024.108083

Cited by 6 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Apart from these, there are many other various applications that have been explored. Autonomous Spacecraft Operations [89] , Intelligent EOS (Earth Observation Systems) [90][91] while another study talks about Autonomous Telemetry systems for Spacecrafts [92] .…”

Section: Previous Studies Of Iot Implementation In Aerospacementioning

confidence: 99%

Applications of Internet of Things (IoT) in the Aerospace Industry

Jain,

Seth,

Singh

et al. 2024

Int. J. Res. Publ. Rev.

View full text Add to dashboard Cite

As we transition to a technology-driven age, it is essential for us to understand the various technologies working behind it, across various industries. The aerospace industry is one such industry that has been massively impacted by the new-age technologies. This study aimed to explore the implementation of one such technology -Internet of Things (IoT) in the Aerospace industry. The study adopted a qualitative, secondary research design which fetched data from published sources (books, news snippets, publications, industry, and company reports) to conduct a thematic and comparative analysis of IoT implementation in the Aerospace industry. The thematic analysis segmented various applications of IoT across 7 segments -Supply Chain, Aero-Manufacturing, ITeS, Research and Development, Avionics, Aero-Operations and MRO (Maintenance, Repair and Overhaul). The comparative analysis studied the variation in applications across the two major sub-industries -Aeronautics and Astronautics. The research conclusively determined the plethora of IoT implementation happening across the segments and sub-industries, while also exploring its challenges and future opportunities.

show abstract

Section: Previous Studies Of Iot Implementation In Aerospacementioning

confidence: 99%

Applications of Internet of Things (IoT) in the Aerospace Industry

Jain,

Seth,

Singh

et al. 2024

Int. J. Res. Publ. Rev.

View full text Add to dashboard Cite

show abstract

“…Their workings and decisionmaking rationale can be more difficult to articulate and understand [17]. Their lack of transparency can be a problem in some applications, such as healthcare [18] and complex aerospace systems [19]. In these application scenarios, the interpretability and accountability of models are extremely critical.…”

Section: Introductionmentioning

confidence: 99%

An interpretable spacecraft flywheel system health status assessment method under perturbation

Zhang,

He,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Health status assessment is an important measure for maintaining the safety of spacecraft flywheel systems. The influence of noise, sensor quality, and other disturbance factors can lead to a decrease in the reliability of the collected information. This can affect the model accuracy. Moreover, a loss of belief in the model is frequently caused by the opaque nature of the procedure and the incomprehensibility of the outcomes, particularly in fields such as aerospace. It is urgent to maintain the interpretability of the model and successfully identify the unreliability of the observed data. Therefore, this paper proposes a spacecraft flywheel system health status assessment method under perturbation based on interpretable belief rule base with attribute reliability (IBRB-r). First, the attribute reliability is calculated based on the average distance method, and a new fusion method of attribute reliability is proposed to reduce the interference of unreliable information. Then, a new interpretable constraint strategy is proposed to improve the rationality and interpretability of the parameters. Finally, the proposed method is validated by a case study of the health status assessment of a spacecraft flywheel system. Experiments show that the IBRB-r maintains high accuracy and interpretability under unreliable observation data.

show abstract