CYBER MONDAY WEEK 🤑 Get 30% off your TNW for Startups or Scaleups packages when you use code CYBER30 only until December 4 →

This article was published on September 18, 2024

Drone swarms edge closer to battle after successful tests in Germany

German Armed Forces commissioned the research


Drone swarms edge closer to battle after successful tests in Germany

A German drone firm claims to have made a “major breakthrough” in autonomous swarm technology.

Munich-based Quantum Systems announced on Tuesday that it has successfully flown AI-controlled Unmanned Aerial Systems (UAS) in swarms.

The company said the AI ensured “reliable mission execution” — even when individual drones completely failed. The system also operated effectively under radio interference.

Commissioned by the German Armed Forces, the tests were conducted last month at the Airbus Drone Centre in Manching, Bavaria.

The 💜 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol' founder Boris, and some questionable AI art. It's free, every week, in your inbox. Sign up now!

The team first trained the AI with deep reinforcement learning — a trial and error approach made famous by DeepMind. This enabled the system to “refine its tactics through continuous self-optimisation,” they said.

After tests in a simulated environment, the AI took to the real skies.

The drone swarm team

The swarm flight united two multi-purpose Airbus drones with the Quantum Systems Vector and Scorpion UAS. As they flew, their reconnaissance data merged in real-time to generate a single picture. This then fed into an Airbus battle management system.

Vector also proved capable of autonomous reconnaissance and target acquisition when GPS is jammed — a big problem for Ukraine’s military drones.

Quantum Systems said the trails had demonstrated AI’s ability to increase UAS resilience. Sven Kruck, the company’s managing director, expects the tech to now rapidly evolve.

“It’s about protecting soldiers and increasing safety,” he said. “In the future, there will be no way around software-based and AI-supported systems for drone technology.” 

Get the TNW newsletter

Get the most important tech news in your inbox each week.

Also tagged with


Published
Back to top