Data mining spacecraft telemetry: towards generic solutions to automatic health monitoring and status characterisation

P. Royer; J. De Ridder; B. Vandenbussche; S. Regibo; R. Huygen; W. De Meester; D. J. Evans; J. Martinez; M. Korte-Stapff

doi:10.1117/12.2231934

15 July 2016 Data mining spacecraft telemetry: towards generic solutions to automatic health monitoring and status characterisation

P. Royer, J. De Ridder, B. Vandenbussche, S. Regibo, R. Huygen, W. De Meester, D. J. Evans, J. Martinez, M. Korte-Stapff

Author Affiliations +

Proceedings Volume 9910, Observatory Operations: Strategies, Processes, and Systems VI; 99101I (2016) https://doi.org/10.1117/12.2231934
Event: SPIE Astronomical Telescopes + Instrumentation, 2016, Edinburgh, United Kingdom

Abstract

We present the first results of a study aimed at finding new and efficient ways to automatically process spacecraft telemetry for automatic health monitoring. The goal is to reduce the load on the flight control team while extending the "checkability" to the entire telemetry database, and provide efficient, robust and more accurate detection of anomalies in near real time. We present a set of effective methods to (a) detect outliers in the telemetry or in its statistical properties, (b) uncover and visualise special properties of the telemetry and (c) detect new behavior. Our results are structured around two main families of solutions. For parameters visiting a restricted set of signal values, i.e. all status parameters and about one third of all the others, we focus on a transition analysis, exploiting properties of Poincare plots. For parameters with an arbitrarily high number of possible signal values, we describe the statistical properties of the signal via its Kernel Density Estimate. We demonstrate that this allows for a generic and dynamic approach of the soft-limit definition. Thanks to a much more accurate description of the signal and of its time evolution, we are more sensitive and more responsive to outliers than the traditional checks against hard limits. Our methods were validated on two years of Venus Express telemetry. They are generic for assisting in health monitoring of any complex system with large amounts of diagnostic sensor data. Not only spacecraft systems but also present-day astronomical observatories can benefit from them.

Citation Download Citation

P. Royer, J. De Ridder, B. Vandenbussche, S. Regibo, R. Huygen, W. De Meester, D. J. Evans, J. Martinez, and M. Korte-Stapff "Data mining spacecraft telemetry: towards generic solutions to automatic health monitoring and status characterisation", Proc. SPIE 9910, Observatory Operations: Strategies, Processes, and Systems VI, 99101I (15 July 2016); https://doi.org/10.1117/12.2231934

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available