AI-based On-board Reconfigurable FDIR and Lifetime Prediction for Constellations

Constellations require complex operations and must grant high service availability. This makes autonomy and enhanced on-board Fault Detection, Isolation and Recovery (FDIR) one of the most urgent developments for the reduction of operational costs and of reaction times to events. This paper presents the results of an ongoing activity aiming at exploiting the assets available during satellites constellations development and operation, such as the models, the Avionics Test Bench, the operational data and events, to develop a workflow that allows fast implementation of high quality AI-based solutions for FDIR and lifetime prediction at constellation level. The ongoing activity benefits from the spin-in of diagnostics and prognostics solutions from similar automotive applications, characterised by combined on-board and off-board processing, operators-in-the-loop and need for on-board solutions easy re-configuration.