Connect with us

Drone Articles

Fly like a bird: Artificial intelligence to assist drones navigating air currents




Story continues from page 1 

The researchers used machine learning to train a glider with a 6 and a half foot wingspan (2 meters) to pick up on environmental cues and navigate atmospheric thermals autonomously. Machine learning is defined by the website techemergence as “The science of getting computers to learn and act like humans do, and improve their learning over time in autonomous fashion, by feeding them data and information in the form of observations and real-world interactions.”

The glider’s instruments could detect changes in wind and make micro adjustments to the bank and angle of the glider. Over many flights and a flight time totaling 16 hours, the autonomous glider grew steadily better at anticipating which turns and angles to take in response to wind conditions and therefore would remain in flight for longer. The researchers results were published the results of their study in the prestigious scientific journal Nature.

Eagles are another type of soaring bird which use thermal currents to stay in he air while expending little energy.

The implications of this research are numerous. If conventional aircraft were to be more efficient at using data from thousands of hours of flying, and using machine learning to better detect thermal updrafts, they could potentially use less energy (fuel) on long journeys which would mean a little less carbon being pumped into our precious atmosphere. Nature continues to show us the way!

For other cool stories we’ve written about drones in nature, check out how drones are helping curb elephant poaching in Africa  or the Top 6 wild animals verses drone encounters.

See all the latest WeTalk videos and articles by following us on social media:

2 of 2Next Page
Use your ← → (arrow) keys to browse


Our Videos