CAMBRIDGE, Mass. — A Massachusetts Institute of Technology (MIT) researcher has developed a system that helps drones maneuver obstacles during its flight, according to MIT’s Computer Science and Artificial Intelligence Lab (CSAIL).
The obstacle-detection system allows drones to “autonomously dip, dart and dive” through tree fields at speeds up to 30 miles per hour, according to a post on CSAIL’s website.
RELATED: U.S. Officials Plan to Require Recreational Drone Owners to Register Aircrafts with Government
“Everyone is building drones these days, but nobody knows how to get them to stop running into things,” said CSAIL PhD student Andrew Barry, developer of the system. “Sensors like lidar are too heavy to put on small aircraft, and creating maps of the environment in advance isn’t practical. If we want drones that can fly quickly and navigate in the real world, we need better, faster algorithms.”
The algorithm builds a full map of the drone’s surroundings in real-time and can detect approaching objects. It operates at 120 frames per second.
RELATED: Drone Attempting to Smuggle Contraband Over Walls of Oklahoma State Penitentiary Crashes
