DARPA Program Finds AI Algorithm to Defeat Human Pilot; Justin Mock, Col. Daniel Javorsek Quoted

2 mins read

During a Defense Advanced Research Projects Agency (DARPA) competition program, an artificial intelligence (AI) algorithm, developed by Heron Systems, defeated a human F-16 fighter pilot in a virtual dogfight simulation, C4ISRNET reported on Friday.

“From a human perspective, from the fighter pilot world, we talked about we trust what works. And what we saw was that in this limited area, in this specific scenario, we’ve got AI that works,” said Justin Mock of DARPA and fighter pilot.

The algorithm easily defeated the fighter pilot in all five rounds, which concluded DARPA’s Air Combat Evolution program competition. The program was launched to analyze automation in air-to-air combat and improve human trust in AI systems.

The conditions in the simulation were not  equivalent to realistic aerial combat. The AI system had perfect information, which experts commentating stated never occurs in the field, and the human pilot flew a fake stick in a virtual seat.

Heron’s AI system exhibited notable aggressiveness and accuracy that would attempt shots that are unlikely to occur in training environments. While the system was successful, it experienced some error in basic fighter maneuvers by turning away from enemy aircraft, but would recover by its “superior aiming ability,” Mock added. 

Heron was one of eight AI teams selected by DARPA to take part in the final round of the agency’s competition. Other competitors included Lockheed Martin, Perspecta Labs, Aurora Flight Sciences, EpiSys Science, Georgia Tech Research Institute, PhysicsAI and SoarTech.

“Artificial intelligence shows a lot of promise. It’s kind of been bang or bust in the past,” Col. Daniel Javorsek said. “In the larger ACE program, our plan is to take the modeling and simulation work that we’re doing here and translate it from that digital environment into the real world. And it turns out that’s a pretty important jump to make.”