Arthur Dominic Villasanta — China Topix June 28, 2016
An artificial intelligence (AI) algorithm named ALPHA developed in the United States has beaten all other U.S. Air Force and U.S. Navy air combat simulation AIs and human fighter pilots in the clearest evidence yet that AIs might one day rule air-to-air combats or dogfights.
Goodbye, Lt. Pete “Maverick” Mitchell, the brash F-14 fighter jock in the 1986 classic hit movie, Top Gun played by Tom Cruise.
Hello, ALPHA, or whatever call sign U.S. fighter pilots will one day give this new Top Gun.
ALPHA’s most famous human “victim” was retired USAF Colonel Gene Lee who has extensive dogfighting experience and is an instructor and Air Battle Manager with a lot of jet fighter aircraft expertise. The AI developed by a University of Cincinnati doctoral graduate beat Lee in a high-fidelity air combat simulator.
Lee, who’s been flying in simulators against AI opponents since the early 1980s, described ALPHA as “the most aggressive, responsive, dynamic and credible AI I’ve seen to date.”
“I was surprised at how aware and reactive it was,” he said of his first dogfight with ALPHA.
“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.”
He noted that with most AIs, “an experienced pilot can beat up on it (the AI) if you know what you’re doing. Sure, you might have gotten shot down once in a while by an AI program when you, as a pilot, were trying something new, but, until now, an AI opponent simply could not keep up with anything like the real pressure and pace of combat-like scenarios.”
ALPHA, however, was specifically designed for use with Unmanned Combat Aerial Vehicles (UCAVs) in simulated air combat missions. It’s viewed as a research tool for manned and unmanned teaming in a simulation environment. In its earliest iterations, ALPHA defeated other AI opponents in simulated dogfights.
“ALPHA is already a deadly opponent to face in these simulated environments,” said Nick Ernest who created ALPHA as a doctoral graduate student at UC.
“The goal is to continue developing ALPHA, to push and extend its capabilities, and perform additional testing against other trained pilots. Fidelity also needs to be increased, which will come in the form of even more realistic aerodynamic and sensor models. ALPHA is fully able to accommodate these additions, and we at Psibernetix look forward to continuing development.”
Ernest later went on to establish Psibernetix, Inc., of which he’s now president and CEO. With him are David Carroll, programming lead; Gene Lee; Kelly Cohen, UC aerospace professor; Tim Arnett, UC aerospace doctoral student and USAF Research Laboratory sponsors.
Lee can attest to ALPHA’s deadliness. It was only after early iterations of ALPHA defeated other AIs that Lee then took to manual controls against a more mature version of ALPHA.
Lee wasn’t able to score a single kill against ALPHA in numerous simulated dogfights. On the other hand, Lee was shot down every time by the AI.
Lee has trained thousands of U.S. fighter pilots over the decades and graduated from the U.S. Fighter Weapons School that’s home to the Top Gun program, and which is where the 1986 Top Gun movie is set.
ALPHA then went on to defeat other human fighter pilots. It even shot down its human opponents despite being deliberately weakened by having its speed reduced; its turn radius constricted and weapons-targeting sensors compromised.