Human pilot beats AI in drone race
Something you are able to do, AI can do higher. Finally.
On October 12, NASA put by itself demonstration, pitting an AI-piloted racing drone in opposition to world-renowned drone pilot Ken Bathroom.
Researchers at NASA’s Jet Propulsion Laboratory, who’ve spent the final two years engaged on drone autonomy (which was funded by Google), constructed three customized drones outfitted with cameras for imaginative and prescient and algorithms that may assist them fly at excessive speeds whereas nonetheless avoiding obstacles.
The drones, named Batman, Joker, and Nightwing, used algorithms that have been built-in with Google’s Tango know-how, which helps AI map out 3D areas.
These drones might fly as much as 80mph in a straight line, however on this significantly cramped course, have been solely in a position to hit 40mph.
In a press launch, NASA defined the professionals and cons of each the autonomous drones and the human pilot. Whereas the AI-powered drones have been in a position to fly extra persistently, they have been additionally extra cautious and, at occasions, bumped into issues with movement blur at larger speeds. Alternatively, Bathroom was in a position to study the course after a couple of laps and fly with way more agility than the autonomous drones, however is prone to fatigue.
“This is definitely the densest track I’ve ever flown,” Bathroom stated within the launch. “One of my faults as a pilot is I get tired easily. When I get mentally fatigued, I start to get lost, even if I’ve flown the course 10 times.”
Lengthy story brief, the AI and the human began out with comparable lap occasions, however Bathroom finally gained out and ended up with a quicker common lap time than the AI.
The implications listed here are massive: autonomous drones might finally be used for surveillance, emergency response, or stock in warehouses.