Drones learn to navigate autonomously by imitating cars and bicycles powered by AI

January 23, 2018, by University of Zurich

Developed by UZH researchers, the algorithm DroNet allows drones to fly completely by themselves through the streets of a city and in indoor environments. Therefore, the algorithm had to learn traffic rules and adapt training examples from cyclists and car drivers.

All today’s commercial drones use GPS, which works fine above building roofs and in high altitudes. But what, when the drones have to navigate autonomously at low altitude among tall buildings or in the dense, unstructured city streets with cars, cyclists or pedestrians suddenly crossing their way? Until now, commercial drones are not able to quickly react to such unforeseen events.

For further info at University of Zurich or IEEE Spectrum Report

Literature:

Antonio Loquercio, Ana Isabel Maqueda, Carlos Roberto del Blanco und Davide Scaramuzza. DroNet: Learning to Fly by Driving. IEEE Robotics and Automation Letters, erscheint am 22. Januar 2018. DOI: 10.1109/LRA.2018.2795643

Video and research site: http://rpg.ifi.uzh.ch/dronet.html