Drones that work autonomously without the need for a human pilot? That certainly sounds like a plan, and the likes of Parrot has gone ahead with its ‘Flight Plan’. The thing is, for drones to get around in the air without hitting an obstacle is not easy at all, as obstacle-detection and motion-planning happen to be extremely tricky subjects where computer science is concerned, no thanks to the creation of real-time flight plans as well as to go up against surprises such as those of wind and weather. Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have demonstrated software which would enable drones to be far more agile and perform hairpin movements in a simulated “forest”, as you can see in the video above.
Imagine a tiny quadrotor performing donuts and figure-eights via an obstacle course of strings and PVC pipes, this particular drone tips the scales at slightly more than an ounce, where it clocks in at 3 and a half inches from rotor to rotor, allowing it to go through the 10-square-foot space at speeds of more than 1 meter per second – all without hitting anything.
Recent graduate Benoit Landry ’14 MNG ’15, shared, “Rather than plan paths based on the number of obstacles in the environment, it’s much more manageable to look at the inverse: the segments of space that are ‘free’ for the drone to travel through. Using free-space segments is a more ‘glass-half-full’ approach that works far better for drones in small, cluttered spaces.”
Filed in . Read more about Drones. Source: scienceblog