An AI-assisted drone has defeated human-controlled drones in an obstacle course measuring precision flight patterns and speed, marking a milestone in mobile robotics.
Humans have been defeated by AI-assisted robots in non-physical games such as chess, checkers, Go, Othello, and StarCraft, but a drone competition staged by researchers at the University of Zurich marks the first time an autonomous drone defeated human pilots in a physical battle.
With the introduction of smaller, quicker drones outfitted with high-resolution cameras, “first-person view” (FPV) has grown in popularity.
RELATED: AI Systems Have Been Discovered To Outperform Humans In The Creation Of Urban Planning Designs
Drone pilot sharpness is being showcased in competitions around the world as the high-speed machines glide through tough maneuvers. The 2023 World Drone Racing Championship will be held next month at South Korea’s Namwon Sports Town facility, with 120 pilots from 30 countries competing.
Zurich University researchers claimed in an article published in Science Robotics that their Swift device defeated three master drone pilots in 15 of 25 tasks.
“Our result marks the first time that a robot powered by AI has beaten a human champion in a real physical sport designed for and by humans,” said Elia Kaufmann, a member of the Swift development team.
Swift uses an artificial neural network to optimize the drone’s course and speed. It collects information in real time using onboard cameras. Human pilots rely on visual feeds sent to headsets to provide a “first-person view.”
RELATED: AI Tools Are Being Employed By Cops Who ‘Do Not Comprehend How These Technologies Work’: Research
The course comprised of seven square poles that drones had to fly through across a 27-year-square field. The goal is to finish the request with no mistakes and in the shortest period of time possible. In addition to defeating human pilots in more than a dozen trials, the Swift drone achieved the quickest speed, although by a fraction of a second.
The task was intimidating. “This task requires pushing the aircraft to its physical limits of speed and acceleration,” according to the paper. Tolerance for error is low: a minor error might result in a catastrophic crash or a significant penalty on lap time.”
The study stated that even little errors made by AI or humans “manifest themselves in reduced task performance, making drone racing a particularly demanding and instructive setting for testing the limits of control design paradigms.”
RELATED: Symphony Secures $1 Million In Seed Funding For Its AI-Powered Music Marketing Platform
Swift’s efficiency decreased when minor adjustments to the course, such as illumination, were altered, according to the researchers.
Smart drones hold great promise in a wide range of applications, including more than just delivering your newest Amazon buy. Farmers can use drones to monitor their crops for growth, disease, and bug infestations. Engineers can use drones to investigate downed power lines or bridge damage, which are dangerous (for humans) yet necessary duties to prevent issues from developing.
A drone routine exploring rubble or flooded areas can aid in search and rescue attempts for victims of natural disasters. Police can use drones to assist in the capture of fugitives.
RELATED: People Are Quite Concerned About The Future After Hearing “Spooky” Audio Of AI Scheduling A Tesla Sales Meeting
In sports, drones are being used to provide previously unachievable aerial views of the event. The PGA Tour also makes use of drone photography.
Drones are currently on the sidelines of such sporting events. Human players, on the other hand, should take care when they are instructed to carry a ball and tackle opponents.
Download The Radiant App To Start Watching!
Web: Watch Now
LGTV™: Download
ROKU™: Download
XBox™: Download
Samsung TV™: Download
Amazon Fire TV™: Download
Android TV™: Download