A researcher at the Massachusetts Institute of Technology has flight-tested a drone that can detect obstacles and avoid them in fight without any input from an operator, MIT said this week. Andrew Barry, a graduate student in the school’s artificial-intelligence lab, has tested the system in a tree-filled field at speeds of about 30 mph. The drone autonomously dips, dives, and changes direction to fly safely through the trees. Barry’s stereo-vision algorithm enables the drone to detect objects and build a full map of its surroundings in real-time. The software, which is open-source and available online, operates at 120 frames per second.
Barry said the system works because he realized that rather than trying to build a full map of the drone’s flight path, the drone really only needs to know what’s about 10 meters away — that gives it enough time to react and avoid the obstacle. “You don’t have to know about anything that’s closer or further than that,” Barry says. “As you fly, you push that 10-meter horizon forward, and, as long as your first 10 meters are clear, you can build a full map of the world around you.” Barry says that he plans to further improve the algorithms so they can work at more than one depth, and in environments as dense as a thick forest. The test drone, which weighs just over a pound and has a 34-inch wingspan, was made from off-the-shelf components costing about $1,700, including a camera on each wing and two processors “no fancier than the ones you’d find on a cellphone,” according to MIT.
By Mary Grady | November 4, 2015