A new artificial intelligence system called ALPHA, developed by Psibernetix, a company founded by University of Cincinnati doctoral graduate Nick Ernest, in collaboration with the U.S. Air Force Research Laboratory, was recently put to the test in fighter plane simulator against U.S. Airforce Colonel Gene “Geno” Lee.
In a series of air combat situations, ALPHA successfully evaded attacks and shot its pursuer down every time. This is impressive not so much because a computer algorithm did better than a human, but because it did better than Lee who is a highly trained professional, having participated in thousands of air-to-air intercepts as mission commander or pilot.
While this is certainly not the first time the Colonel crossed swords with an A.I. – actually, he’s been doing this for decades – he found ALPHA to be, by far, the best digital opponent he’s ever encountered.
“I was surprised at how aware and reactive it was. It seemed to be aware of my intentions and reacting instantly to my changes in flight and my missile deployment. It knew how to defeat the shot I was taking. It moved instantly between defensive and offensive actions as needed,” said Lee.
ALPHA, specifically designed for research purposes in simulated air-combat missions, uses a decision-making system called a genetic fuzzy tree (GFT), which is a subtype of fuzzy logic algorithms. The system approaches complex problems much like a human would by breaking a larger task down into smaller subunits, which include high-level tactics, firing, evasion, and defence.
This enables the system to consider only the most relevant variables and come up with an optimal maneuver as many as 250 times faster than a human ever could. While the researchers admit that the human mind will probably always outperform A.I. in certain areas, it is simply no match for it here.
Facing off with such an advanced system in hour-long combat missions left Colonel Lee beyond tired: “I go home feeling washed out. I’m tired, drained and mentally exhausted. This may be artificial intelligence, but it represents a real challenge”.
Further applications of GFT methodology could include robotic surgery, design automation, cyber security and many others.
Results of the experiment were published in the Journal of Defense Management.