Raspberry Pi-powered AI solution defeats US Air Force pilot in combat simulation

While the original aim of launching the Raspberry Pi was to just let enthusiasts test their DIY projects, now, leveraging some software tweaks, it can beat even humans at their own game. Nick Ernest, a doctoral graduate from the University of Cincinnati, has designed an artificial intelligence (AI) solution for the single-board computer to prove its capabilities.

Called ALPHA, the AI solution defeated retired US Air Force Colonel Gene Lee during combat simulation. Lee described the new development as “the most aggressive, responsive, dynamic and credible AI” solution, to date.

Flying against opponents on a flight simulator was not a new task for Lee as he has been beating computer programs since the early 1980s. However, ALPHA made the pilot “tired, drained and mentally exhausted” due to its exceptional results. “I was surprised at how aware and reactive it was. It seemed to be aware of my intentions and reacted instantly to my changes in flight and my missile deployment. It knew how to defeat the shot I was taking,” Lee said.

Ernest aims to enhance the AI program in the future by extending its capabilities and reducing mistakes. “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 explained.

ALPHA could become the ultimate solution for air combat simulation. It could even emerge as a benchmark for other AI programs, and perform tasks based on the commands of a manned wingman in actual flight, using the power of Raspberry Pi.