Autonomous cars haven't even fully hit the roads yet, and companies are already touting the potential benefits of autonomous drones in the sky – from package delivery and industrial inspection, all the way to modern warfare. But a drone presents new levels of challenges beyond a car. While a self-driving car or land-based autonomous robot at least has the ground underneath it to use as a baseline, drones potentially will have a full 360-degree space to move around in and must avoid all of the inherent obstacles and pitfalls associated with this. Speaking at the 2017 GPU Technology Conference (GTC), a team of engineers from Nvidia believe the solution to having freely autonomous drones lies in deep learning. Their research has already yielded a fully autonomous drone flight through a 1 km forest path while traveling at 3 m/s, the first flight of its kind according to Nvidia.

A U.S. Air Force pilot grasps a flight control and weapons firing stick while preparing to launch a MQ-1B Predator unmanned aerial vehicle (UAV), from a ground control station in the Persian Gulf region on January 7, 2016. At 12:01 p.m. on January 20th, drone geek Maynard Holliday won't be a Pentagon robot expert anymore. One of the roughly 4,000 Obama appointees out of a job on Friday, Holliday is wrapping up two and a half years of work designing and buying drones for the Pentagon. A self-proclaimed Star Trek-loving science nerd who grew up obsessed with the thought of one day traveling into space himself, today Holliday's much more interested in a smaller kind of'space race' happening here on Planet Earth. As he shuffles his way out of Washington, he says he's got one big worry about the future of military drone warfare: What happens when and if killer drones go on autopilot?