A drone detector that says
"I don't know this one."
Angel Eyes listens to the radio link between a drone and its controller, identifies the drone when it recognises it, and refuses to guess when it does not. It detects and hands off. It does not jam, spoof, or engage.
Why refusing to guess matters
Most classifiers are built to always answer. Shown a drone they have never seen, they return the closest thing they know, with high confidence. In counter-drone work that is the dangerous failure: a confident wrong name is worse than an honest "unknown", because it ends the investigation instead of starting one.
Angel Eyes is built and measured around that second case - the drone that is not in the catalogue.
Where it stands today
Prototype, not a product. The numbers below are real and measured, on public
data, on one split. They have not been replicated across other splits, and nothing
here has been tested against live off-air signals. Read
Limits before drawing conclusions.
77.1%of never-before-seen drones correctly refused
84.2%of known drones correctly identified
0.915separability, known vs unknown (AUROC)
Single split (seed 42), public RFUAV dataset, not yet replicated.
Scope, stated plainly
- Does: detect a drone control link, identify the model when known, refuse when unknown, hand the result to partner hardware.
- Does not: jam, spoof, take over, cue a weapon, or intercept. Not now, not planned.