How drones learn to find their way around big cities
An algorithm developed by researchers at the Swiss Federal Institute of Technology (ETH) Zurich aims to help drones practice safe behavior in complex urban scenarios.
Scientists at the University of Zurich have christened the algorithm that will in future enable safe drone flights in densely populated and heavily frequented environments DroNet . The unmanned autonomous aircraft will apparently learn from the data collected during car and bicycle rides through the cities. In doing so, they pick up both traffic rules and reaction patterns that allow them to face unknown situations, obstacles and other risks.