Machine Learning-based vs Deep Learning-based Anomaly Detection in Multivariate Time Series for Spacecraft Attitude Sensors

TL;DR

This research characterizes two different approaches to the problem of stuck values detection in multivariate time series coming from spacecraft attitude sensors, revealing the performance differences in the two approaches.

Abstract

In the framework of Failure Detection, Isolation and Recovery (FDIR) on spacecraft, new AI-based approaches are emerging in the state of the art to overcome the limitations commonly imposed by traditional threshold checking. The present research aims at characterizing two different approaches to the problem of stuck values detection in multivariate time series coming from spacecraft attitude sensors. The analysis reveals the performance differences in the two approaches, while commenting on their interpretability and generalization to different scenarios.

Figures (2)

• Figure 1: Example of a stuck at last value on all three axes of the accelerometer
• Figure 2: Example of a stuck at last value on all three axes of the IMU