Competing drone control technology: humans are faster and AI operation is more stable

On November 24, according to CNBC, drone competition is a booming industry. So who can be a better drone operator? Robot or human? NASA conducted this test on October 12 and announced the results today: World-class drone pilots are faster, but AI-operated drones are more stable. The NASA team introduced drone pilot Ken Loo to compete with artificial intelligence (AI)-controlled drones. The drone drove an average of 11.1 seconds to complete the game, while the AI-controlled drone required an average of 13.9 seconds. But this game is tricky. Lo said in the statement: “This is definitely the densest track I’ve ever flown in. As a pilot, one of my shortcomings is that I easily fatigued. When I was mentally exhausted, I started to get lost, even if I was already This track has flown 10 times." NASA said that in fact, AI-controlled drones are more stable. Rob Reid, task manager of the project, said in a written statement: "Our algorithm is aimed at humans, and the more unstable human performance is, the more unstable you can see. You can see that AI lets drones. Smooth and steady flight on the track, while human pilots tend to accelerate aggressively, so their flight path is unstable." The drone used in the competition was created by researchers from the NASA-sponsored Jet Propulsion Laboratory. Google funded NASA's two-year unmanned aerial vehicle research, and this competition is an outcome test. Of course, developing drone technology is not just for competition. Such drones will be able to check inventory in warehouses, aid in disaster search and rescue operations, and possibly help robots navigate in space stations. The NASA team built three drones, named them Batman, Clown and Nightwing, and programmed them by algorithms. For this reason, these drones can be quickly made without hitting objects. flight. Reid said: "Just one day, you may see their professional competition!" (Small)

Posted on