Raspberry Pi AI Beats Human Fighter Pilot in Simulation

Raspberry Pi AI Beats Human Fighter Pilot in Simulation

An experienced former US Air Force officer was beaten in a simulated fighter plane dogfight against an artificial intelligence running on a Raspberry Pi. ALPHA – the AI against which retired Colonel Gene Lee went head-to-circuit-board – was developed by a doctoral graduate from the University of Cincinnati. But Colonel Lee wasn’t the only loser during combat simulations: the Air Force Research Lab’s own AIs – every single one – also fell to the might of ALPHA.

Lee has decades-worth of flight time under his belt, experience which he later used as an instructor to train thousands of young fighter pilots, and yet he failed to bag even one victory against ALPHA.

“I was surprised at how aware and reactive it was,” Lee said (via Science Alert). “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.”

ALPHA’s unbeatable ability stems from its genetic fuzzy tree system – a fuzzy logic algorithm subtype – which it uses to calculate strategies and counter-strategies 205-times faster than a human blink.

“ALPHA could continuously determine the optimal ways to perform tasks commanded by its manned wingman, as well as provide tactical and situational advice to the rest of its flight,” Kelly Cohen, UC aerospace professor, said.

The success of ALPHA is sure to stir up ethical issues regarding the use of artificial intelligence against human opponents in real combat situations.

“ALPHA is already a deadly opponent to face in these simulated environments,” said Nick Ernest, founder of ALPHA developer Psibernetix. “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,” Ernest added. “ALPHA is fully able to accommodate these additions, and we at Psibernetix look forward to continuing development.”


Popular posts from this blog

"Investors need a minimum." When Bitcoin Will Begin Recovery

When Bitcoin Will Be More Expensive Than $100 Thousand: A Selection of Predictions

In the US, launched mining at a nuclear power plant