This Tuesday, specialists from NASA’s Jet Propulsion Laboratory (JPL) in Pasadena, CA released data showing that autonomous drones performed better than human-controlled drones. This indoor race course consisted of various obstacles for the drones to work around as observers took note of the advantages and weaknesses of both human and AI contestants. Although autonomous drones were slightly slower than their mortal counterparts, they were consistently efficient and didn’t become fatigued or mentally unfocused. This recent disclosure signifies the end of a two-year research project funded by Google aimed at better understanding JPL’s vision-based navigation in drone technology.
With wide field cameras on both the top and bottom of its structure, the drones have a built-in map of the course to navigate its environment accurately. These internal algorithms were developed by Google’s AR platform, Tango. The three customized drones named Batman, Joker, and Nightwing hit speeds of 30-40 mph maneuvering what resembled a futuristic hot wheels race track. A self-driving drone went head to head with Google’s senior product design engineer and professional drone pilot, Kevin Loo AKA “Flying Bear.” Loo was able to see his drone’s outlook by wearing first-person view goggles averaging 11.1 seconds per lap compared to AI’s average of 13.9 seconds. Initially taking place on October 12, 2017, Loo admitted the challenges he encountered during the competition.
This is definitely the densest track I’ve ever flown. 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.
JPL’s project task manager, Rob Reid, added to Loo’s statement observing that:
You can actually see that the A.I. flies the drone smoothly around the course, whereas human pilots tend to accelerate aggressively, so their path is jerkier.
(Courtesy of NASA/JPL)
However, Reid acknowledged that Google’s algorithms are still a work in progress. For instance, they found the drone’s vision becoming blurred by moving too quickly. Additionally, GPS networks are extremely limited indoors posing a significant challenge for autonomous technology. Still, Reid remains optimistic that this is just the beginning of the vast opportunities that come with self-driving drones. They could monitor inventory or conduct search and rescue missions during times of emergency. They also might be used to assist robotic technology “navigate the corridors of a space station.” Easily hitting speeds of 80 mph when going in a straight line, these drones could also be a source of entertainment similar to the competition held in JPL’s warehouse.
Our autonomous drones can fly much faster, Reid said, One day you might see them racing professionally!
There’s no doubt that autonomous technology will be integrated into tomorrow’s landscape, but its application in society is still up for debate. Emotions are a human characteristic that makes us unique, but AI has the advantage of not letting feelings get in the way of tasks at hand. Moreover, this technology is only going to become more sophisticated in hopes of one day surpassing organic intelligence.
What do you guys think of the latest drone race? Does this show that AI will eventually become superior to man or do you think that will never happen? Drop a note and let us know what you think!
Remember to subscribe to our newsletter!
Follow The Scope Weekly on Twitter and Facebook and Instagram!
If you would like to become a contributor to The Scope Weekly, read our submission guidelines, and apply. For product reviews, click here.