MIT algorithm could help prevent midair collisions
System out of MIT International Center for Air Transportation creates puck-shaped GPS zones around small aircraft to prevent collisions and near-misses.
MIT researchers are working on an algorithm that could help reduce the likelihood of airplane collisions in the sky, part of work to overhaul the FAA air traffic system.
The FAA's NextGen overhaul mandates that by 2020 all commercial aircraft broadcast GPS coordinates, which would be more accurate than ground-based radar.
The system uses GPS data to track hypothetical puck-shaped zones around smaller aircraft to keep them a safe distance apart. Thousands of small aircraft were involved in near-misses over the past decade and there were 112 midair collisions, according to MIT.
Researchers at the MIT International Center for Air Transportation (ICAT) based the system on months of real-world flight data. A chief goal was to reduce the frequency of false collision alarms. They decided to use two alerts: a moderate one when flight paths are converging, and a severe alert when a collision is imminent.
The size of the puck-shaped warning zone fluctuates according to trajectory and time remaining before a potential crash. The system had a low incidence of false alarms when tested in a computer model based on real-world airport data from across the U.S., according to MIT.
One problem, however, is that the model uses radar data, and many small aircraft fly below radar around airports. The ICAT researchers are going to have to show that their model can also work with typical flight paths used by small aircraft.
The researchers plan to present their work at the 30th Digital Avionics Systems Conference (DASC) in Seattle in October. They also want to start testing it on real planes. That's not an experiment I would want to join.