AI system beats US military pilot in F-16 dogfight

An artificial intelligence (AI) system has beaten a US military pilot in a five-round challenge to demonstrate whether autonomous computer systems could be capable of flying fighter jets.

America’s Defence Advanced Research Projects Agency (DARPA) ran its final AlphaDogfight trial this week as part of a programme to see whether autonomous systems could defeat adversary aircraft in a simulated dogfight.

It’s a precursor to DARPA’s air combat evolution (ACE) programme, which will eventually fly live aircraft using AI algorithms to demonstrate that the technology will work in the real world.

One of the AlphaDogfight simulated battles. Pic: DARPA
Image: One of the AlphaDogfight simulated battles. Pic: DARPA

On Thursday, the AI developed by Heron Systems defeated an experienced F-16 pilot in each of the five rounds – although the success in a simulation is just one step towards fully autonomous systems being used on F-16 fighter jets.

The moment is not quite as seminal as when IBM’s computer Deep Blue defeated the former world chess champion Garry Kasparov in 1997, effectively confirming that computers were capable of playing chess better than humans.

Chess is a game with a finite number of possible moves, and accommodating all possible moves to defeat any human opponent has proven to be well within the ability of a modern computer.

Although the computer simulation for the dogfight programme is much more complicated than the game of chess, it is similarly constrained by the number of moves and variables which have already been coded into the simulation – and there are no such constraints on dogfights in real life.

More from Science & Tech

The simulation also presented a Second World War-style scenario for the planes, where the aerial battle was based on the idea that the fighter aircraft had only forward-facing guns to attack each other with.

Modern dogfights can be much more complex, especially when they include missiles, which were not used during the simulations.

US Air Force Lieutenant Colonel Dan Javorsek, who is DARPA’s ACE programme manager, was cautious about the limitations of the simulation in proving the AI’s superiority to a human pilot.

Lt Col Javorsek said: “Just because the outcome went the way it did today it does not mean that the AI wasn’t latching on to some weird digital artefact that we weren’t thinking about or considering, and that’s what was giving them an advantage.”

“There were certainly some peculiar behaviours in the engagement,” he added, before stressing that the result was still a human triumph, as humans developed Heron’s successful AI system.

Let’s block ads! (Why?)

Tech News – Latest Technology and Gadget News | Sky News

Leave a Reply

Your email address will not be published. Required fields are marked *