Autonomous Algorithm Enables Drones to Outrace Human Pilots
When I first saw the headline about a drone beating a human pilot in an air race I was shocked or intrigued perhaps. How on earth would they even set that up? Did they give an algorithm an airplane to fly? Isn’t that dangerous? After reading a little more of the gizodo.com article I found it was actually two human pilots. But AI was flying a drone and the humans were flying a similar drone. And AI won.
It’s not likely full-sized planes could ever do what these drones were asked to do. The article by Andrew Liszewski gave some insight as to how the drone race was run.
Unlike auto racing, in drone racing, the pilots aren’t subjected to debilitating G-forces. That means the races can take place at impossibly fast speeds. So fast, in fact, that autonomous drones have always lagged behind those piloted by humans who benefit from split-second reflexes. But that’s no longer the case.
Researchers at the University of Zurich have developed a new algorithm that can analyze a course with specific waypoints. Drone racing isn’t just flying in circles like Nascar—it requires the craft to navigate a complex layout of obstacles that often require tight turns or complete direction changes. The algorithm then calculates the most efficient route for a drone to fly, including the speeds, angles, and trajectories at which it navigates around or through the obstacles.
Basically, after the algorithm memorized certain views of the course- when it sees those views again, it was full speed ahead–and enough to smoke the human pilots on the same course with the same prep time. Take a look below.
Yes, the AI had an advantage, as all AI does. However, researchers feel these algorithms will be used for things like delivering medications safely or even food items. There is still a lot of work to be done on battery strength and other areas that can be improved, but sports fans may be able to have a beer delivered to them at an event soon.
read more at gizmodo.com