A network that teaches drones to navigate autonomously

With the expansion of UAVs and their uses, the necessity of safely flying them through the urban environment without accidents has become evident.

The scientists at University of Zurich and the National Centre of Competence in Research (NCCR) Robotics are working on a network that teaches the drones to autonomously navigate through cities avoiding collisions.

DroNet (short for Drone Network) is a convolutional neural network (CNN) that can safely fly drones through a city at low altitudes by imitating how cars and bicycles navigate. The navigation policy learned by DroNet is highly generalizable and will even teach the drones to navigate autonomously through parking lots, corridors, and other indoor environments.