Version 11 Of The AirBurr UAV Learnes From Its Crashes



future, AirBurr UAV, EPFL, robot concept, robotics, Laboratory of Intelligent Systems, future robot, UAV, EPFL, futuristic
Having studied the way a fly move around a house, bouncing off a lot of windows and walls, until it finds a route, researchers at Switzerland’s EPFL Laboratory of Intelligent Systems designed the latest version of their insect-inspired autonomous AirBurr UAV that runs into things, to be able to map and navigate its environment. Like its forerunners, Version 11has rotor blades, electronics and other delicate bits housed within an open, flexible carbon fiber frame offering protection in case of a crash. When the UAV does get knocked down, it’s able to right itself by extending four retractable carbon fiber legs from its sides and continue its flight. New is its ability to learn from its crashes: analyzing the position and force of its collisions, the AirBurr can gradually map out its surroundings, establishing where the various boundaries lie. The UAV’s onboard sensors consist of just four photodiodes, to help when seeking out light sources. Thanks to the bump-and-crash method, this UAV can easily get around pitch-dark environments and this feature could be of exceptional importance in the exploration of cramped and/or dark places such as caves, collapsed mines, or damaged nuclear power plants.
Via:gizmag.com

future, AirBurr UAV, EPFL, robot concept, robotics, Laboratory of Intelligent Systems, future robot, UAV, EPFL, futuristic

future, AirBurr UAV, EPFL, robot concept, robotics, Laboratory of Intelligent Systems, future robot, UAV, EPFL, futuristic

More Posts:

The Future Of Shopping
Jets With Engines On Top Have Better Lift & Aeroacoustics
Goo: Plants Will Grow In Desert And Space
Grabit To Finger Lock Doors
Sustainable Coastal Scallion Skyscraper
Finger Control Patent For Google Glass
Progress Of The NASA's James Webb Space Telescope Program In 2013 (+VIDEO)
The Future Of Collaboration (VIDEO)
IBM Aims For Commercializing Its First Carbon Nanotube Transistors In The Early 2020s
Stanford Team Creates Computer Vision Algorithm That Can Describe Photos