Artificial intelligence (AI) won simulated dogfights against a human F-16 pilot, 5-0; what does that mean?

Aug. 26, 2020
The next steps for the ACE program include integrating an AI agent into a small-scale combat aircraft and eventually to a full-size manned fighter.

PATUXENT RIVER NAS, Md. – The U.S. military experiment called the AlphaDogfight simulation wrapped up last week. On the face of it, man lost badly to machine. In fact, the human F-16 jet fighter pilot, call sign “Banger,” lost all five simulated dogfights to an artificial intelligence agent (AI) from Heron Systems Inc. in California, Md. Forbes reports. Continue reading original article

The Military & Aerospace Electronics take:

26 Aug. 2020 -- Heron systems develops autonomous agents and multi-agent systems powered by AI. The company's system first battled its way past competing AI systems from Boeing, Aurora Flight Sciences, EpiSys Science, Georgia Tech Research Institute, Lockheed Martin, Perspecta Labs, PhysicsAI, and SoarTech.

U.S. Air Force Col. Dan Javorsek, head of the U.S. Air Force Air Combat Evolution (ACE) program, balanced the apparent dogfight success of the AI agent against the limitations of the simulation to add perspective to what appeared to be a straightforward AI victory.

The F-16 jet fighter pilot said the exercise led him to trust the dogfighting algorithms, particularly their demonstrated maneuvering, their rapid decision-making ability, and their targeting capability - but with caveats.

Related: Military researchers to develop artificial intelligence (AI) algorithms for future unmanned jet fighters

Related: Researchers choose SoarTech to develop artificial intelligence (AI) for manned and unmanned dogfighting

Related: DARPA eyes heavier reliance on artificial intelligence (AI) and unmanned aircraft in dogfighting

John Keller, chief editor
Military & Aerospace Electronics

Voice your opinion!

To join the conversation, and become an exclusive member of Military Aerospace, create an account today!