The significance of an MIT drone weaving around tree branches at 30 mph
To get his PhD, MIT grad student Andy Barry packed up a car with a drone and a catapult to launch it. Then he headed west.
Barry’s destination was a farm in western Massachusetts. It had a great restaurant, a friendly owner and a ton of space. Boston, with all of its density, wasn’t the right place to test an automated drone flying at 30 mph.
Barry’s trips to the farm grew out of conversations five years ago with his adviser at MIT’s Computer Science and Artificial Intelligence Lab. Could a lightweight drone be launched at obstacles, and avoid them with no prior knowledge of its surroundings?
Barry didn’t think he had a chance. Computers were too slow. But his adviser reminded him of Moore’s Law, the observation that computing power doubles roughly every two years.
Fast forward to today. Moore’s Law delivered, and now Barry will be defending his thesis next month. A new video shows his drone avoiding trees at 30 mph. There’s nothing special about the components. The drone is powered by the same chip that’s in a Samsung Galaxy S3. Barry shot the video as proof for his thesis committee.