Robust Multimodal FOD Characterization

Foreign Object Damage (FOD) is a leading cause of Class-A mishaps within the aviation community and a major driver of maintenance and repair costs for the Navy.  It is not possible to completely eliminate FOD events, but the Navy would like to better understand FOD risks so that they can be mitigated.  A critical aspect of understanding FOD events is the collection of dynamic data on the debris location, trajectory, size and composition that can be correlated with sensor data collected by the engine to determine the overall effect.  This can be accomplished with an inlet debris monitoring system (IDMS), but existing single-sensor IDMS units are prone to false alarms and have limited sensitivity to certain length scales.  Luna proposes to improve the Navy’s ability to understand FOD events and their risks through development of a versatile, robust, and intelligent multimodal sensor suite that leverages multi-spectral, close range mm-wave radar and optical scattering techniques to sense debris characteristics over a range of length scales, with a robust, embedded machine learning framework for fusing measurement data and engine diagnostics to provide a complete picture FOD events.