CHANNELS

Aerospace

Automotive

Automated & Autonomous Vehicles

Battery & Electrification Technology

Defense

Drone TV

Electronics

Manufacturing & Prototyping

Materials

Medical

Motion Control

RF & Microwave Technology

Robotics & Automation

Sensors & IoT

Test & Measurement

Aerospace
Automotive
Automated & Autonomous Vehicles
Battery & Electrification Technology
Defense
Drone TV
Electronics
Manufacturing & Prototyping
Materials
Medical
Motion Control
RF & Microwave Technology
Robotics & Automation
Sensors & IoT
Test & Measurement
Aerospace & Defense
Search Results

NASA and Google's Drone Race: Human vs. Artificial Intelligence

Engineers from NASA's Jet Propulsion Laboratory  put together a drone race to find which is faster - a drone operated by a human or one operated by artificial intelligence. The race capped two years of research into drone autonomy funded by Google. The company was interested in JPL's work with vision-based navigation for spacecraft - technologies that can also be applied to drones. JPL set up a timed trial between their A.I. and world-class drone pilot Ken Loo. The team built three custom drones and developed the complex algorithms the drones needed to fly at high speeds while avoiding obstacles. These algorithms were integrated with Google's Tango technology, which JPL also worked on. For the official laps, Loo averaged 11.1 seconds, compared to the autonomous drones, which averaged 13.9 seconds. Although, the autonomous drones were more consistent overall - where Loo's times varied, the A.I was able to fly the same racing line every lap.

Topics:
Aerospace Automation Computers Electronics & Computers Imaging Motion Control Robotics Software

More From SAE Media Group

    Transcript

    00:00:01 Human vs. Machine JPL engineers recently built a new kind of drone --one piloted by artificial intelligence. The drone uses cameras to track its position [Engineer's voice] A wide field camera on the front here-- and then also on the bottom here there's a second wide field view camera. [on-screen] then matches it with a pre-loaded map. To test it, engineers set up an obstacle course in a JPL warehouse.

    00:00:31 [animation traces the path of the obstacle course] And just for fun, invited world-class drone pilot Ken Loo to race against it. A.I. driven Human driven [split screen with race-timing counters] The winner? The A.I.-piloted drone averaged 13.9 seconds per lap. Loo's drone averaged 11.1 seconds. Although slower, the A.I.-piloted drone

    00:01:16 completed the course with accuracy --capping off two years of research into drone autonomy funded by Google Don't be surprised if you hear more about autonomous drones in the future. Researchers hope that one day drones could be piloting themselves around warehouses or even navigating disaster areas to find survivors. [drone sounds] NASA Jet Propulsion Laboratory

    00:01:47 California Institute of Technology