NATO Tests Drone Defense Network and Robotic Resupply in Latvia

A sensor detecting a drone is useful only if its data reaches the right operator, crosses national command systems and produces an authorized response before the threat completes its mission. At Latvia’s Ādaži military base, NATO tested that entire chain by connecting sensors, command software, electronic jammers, interceptor drones and ground robots during a simulated attack on the Baltic states.

The experiment’s central premise was that no single defensive system can handle large numbers of inexpensive drones. Instead, radar and other sensor feeds moved into a shared situational-awareness display, allowing command software to coordinate several defensive layers. Unmanned ground vehicles simultaneously carried supplies to friendly units, testing whether forces could maintain logistics while personnel faced persistent airborne threats.

The event brought together three NATO modernization efforts: its Digital Backbone, Layered Counter-Uncrewed Aircraft Initiative and Next Generation Targeting. NATO Allied Command Transformation describes the Digital Backbone as the software, networks and infrastructure needed to connect forces and exchange data. In Latvia, it served as the common transport layer for information produced by counter-drone sensors and targeting applications.

A layered system rather than one defensive device

During the demonstration, incoming drones were addressed with a mix of electronic disruption, interceptor aircraft and a robotic weapon system. Some interceptors were automated and others were flown by human pilots. The defenses prevented the initial drones from completing their simulated missions, although not every aircraft was destroyed.

That result illustrates the logic of a layered defense. An electronic system may disrupt one aircraft, while an interceptor handles another and a command node assigns priorities across the network. The important engineering task is not merely adding more equipment; it is ensuring that sensors, software and response systems from different nations and manufacturers can exchange usable information quickly and consistently.

NATO is expanding this work through its layered counter-drone experimentation. The initiative connects groups of sensors, command nodes and defensive systems so they can share information beyond an individual local cell. More than 100 NATO, allied and industry entities from over 30 allied and partner countries are expected to participate in the wider experimentation effort.

Scaling that architecture presents a harder problem than a successful local demonstration. More connected sensors generate more tracks and data, placing additional demand on communications links and operators. National equipment may describe threats, tasks and priorities differently. Network degradation or incompatible data formats can delay decisions even when every individual component works as designed.

AI bridges systems, but people retain authority

NATO used AI agents to connect national targeting applications that were not originally designed to communicate. The goal was to produce a common operational picture and reduce the manual work required to move information between systems. In one simulated sequence, software reviewed intelligence and assembled a proposed response plan within 60 seconds. Admiral Pierre Vandier, NATO’s supreme allied commander transformation, retained authority to approve the task.

That human authorization is a defined part of the design rather than an optional final check. NATO’s Next Generation Targeting program is testing AI-supported workflows against a baseline without AI, measuring decision time, accuracy, operator workload and decision quality. NATO says trained personnel can intervene throughout the process and a person makes the decision to engage.

The same program is developing a common data model rather than replacing every national command system with one NATO application. Participating countries can retain their own equipment while using shared descriptions for targets, intent and tasks. That approach reduces the need for a wholesale hardware replacement, but it makes standards compliance, software integration and reliable data translation central to readiness.

Robotic resupply extended the experiment beyond interception. Ground vehicles carried supplies to units while the counter-drone network operated, demonstrating how automation could reduce personnel exposure and preserve logistics in an airspace crowded with small unmanned aircraft. Fielding such machines at scale would still require dependable communications, maintainable hardware and procedures for operating around troops and conventional vehicles.

The Latvia experiment did not establish alliance-wide readiness. Its organizers expect testing failures to identify changes needed in interfaces, architectures and operating procedures before broader fielding. Vandier’s warning about potentially enormous drone numbers makes scale the unresolved issue: NATO has demonstrated that separate defensive and robotic systems can be connected, but it must now show that the network remains coherent across nations, manufacturers and disrupted communications.

By Stephen Wallace — Editor for AMI’s aerospace integration and unmanned mobility coverage, focused on drone manufacturing, VTOL systems, autonomous networks, and air-ground mobility links.