Powerus unveils Guardian-3 fully autonomous drone interceptor

The counter-drone market has spent the better part of a decade solving the detection problem. Radar, RF sensing, EO-IR turrets, and acoustic arrays now routinely cue on small unmanned systems at tactically useful ranges. What has lagged is the response loop: the time between a confirmed track and an effective interception, and the human labor required to close it. Powerus's announcement of the Guardian-3, pitched as a fully autonomous interceptor, is best read as a claim about that loop rather than about any single airframe. The company is arguing, in effect, that the binding constraint on counter-UAS is no longer sensing but decision latency and operator throughput, and that the answer is to let the machine complete the kill chain on its own. That framing is the news. The hardware is the vehicle for it.

The mechanics matter because autonomy in this domain is layered. A fully autonomous interceptor implies onboard or edge-hosted perception that can classify a track as hostile, maintain it through clutter and maneuver, and commit to a intercept geometry without a human in the immediate loop. That is a materially different proposition from the assisted systems that dominate current fielding, where a remote operator authorizes engagement. The engineering burden sits in three places: robust classification to avoid false positives against birds, friendly UAS, and civilian traffic; guidance that closes on small, agile targets with limited onboard sensing; and a safety architecture that can demonstrate bounded behavior when the autonomy is uncertain. Powerus has not, in the excerpt, detailed which of these it has solved or how. That gap is where evaluators will focus, because the difference between a compelling demo and a certifiable system usually lives in the third item, not the first two.