Orbit Logic // Fault Learning Agent for Prediction, Protection and Early Response (FLAPPER)

ORBIT LOGIC // FAULT LEARNING AGENT FOR PREDICTION, PROTECTION AND EARLY RESPONSE (FLAPPER)

VIRTUAL BOOTH

Solution Name: Fault Learning Agent for Prediction, Protection and Early Response (FLAPPER)

Challenge: Space Asset Resiliency

Focus Area: Other

View Challenge Submission

You will need to make an AFWERX Challenge account to view.

Solution Description:

The FLAPPER solution will be implemented as a Specialized Autonomous Planning Agent (SAPA) that expands our onboard Autonomous Planning System (APS) architecture to include Machine Learning (ML) capable of detecting, isolating, and mitigating anomalies in real- or near-real-time with minimal ground intervention. The system will utilize spacecraft telemetry to train the ML models to detect faults and link the corresponding responses based on novel telemetry limits, value trends and anomalous behavior.