Autonomous strike drone uses Nvidia Jetson Orin Nano to independently pick and bomb targets — Swedish startup's attack drones run small AI model, require no human input and zero external comms
Drones built with small, non-frontier computer-vision models autonomously identified and attacked targets in a recent demo.
Scaleout Systems, a Swedish AI startup, demonstrated autonomous strike drones in a recent display. The drones utilized low-cost loitering munitions from BAE Systems Bofors to target and detonate, as reported by Ars Technica. The drones autonomously detected, geolocated, and prioritized an armored engineering vehicle, then flew to and struck it without any human intervention.
The entire mission, including reconnaissance, took less than a minute, and the drones could operate without any external communication, making them resilient to electronic warfare. The project, FEDAIR, is part of NATO's DIANA accelerator, receiving funding for development, training, and testing. Scaleout also showcased an "arctic strike demonstration" at a Swedish air base in sub-zero temperatures, where the drone ran an object-detection model on Nvidia's Jetson Orin Nano.
The drone maintained a frame rate of 30 fps and a target latency of 30 ms or less under these harsh conditions. The demonstrations, while not yet leading to official NATO procurement, highlight the potential of autonomous target selection running directly on the drone's airframe.
Written by urgent.news from Tom's Hardware's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
