Scaleout Systems, a NATO-backed startup, is deploying AI-driven target detection and selection designed to run on small drones for surveillance and attack missions. Originally focused on machine learning for commercial vehicles, the Swedish company pivoted to defense applications following the 2022 invasion of Ukraine.
NATO-backed startup Scaleout Systems deploys edge AI for autonomous drone missions
Unlike large-scale frontier models, Scaleout utilizes leaner machine learning models capable of performing computer vision tasks directly on drone hardware or embedded devices at forward bases. This edge computing approach allows drones to operate independently of continuous communication with a central server, providing a strategic advantage in environments where electronic warfare or jamming might disrupt signals.
Through the Federated Aerial Intelligence for Recon project, the company works on adapting models for edge hardware used in drones, pilot tablets, and command posts. These models can identify targets and share selective updates with local command nodes without transmitting sensitive raw data. These nodes can then use the aggregated data to retrain the models and push updated capabilities back to edge devices, a process known as federated learning.
Scaleout is also participating in the Affordable Loitering Modular Ammunition (ALMA) project led by BAE Systems Bofors to develop low-cost, autonomous kamikaze drones. In demonstrations, the system has shown the ability to detect, identify, and geolocate threats in real-time using dedicated onboard computing, allowing drones to prioritize and strike targets without direct human commands.
Sources
- Small AI models let drones autonomously identify and attack battlefield targets （Ars Technica AI, 2026-09-17）