A new algorithm can pilot a drone to guide flocks from away from busy airspace.
The current method: Birds pose a serious danger to planes—just ask pilot Chelsey Sullenberger and the survivors of US Airways flight 1549 in 2009. Airports employ all sorts of tactics to scare birds off, from loud noises to lasers, and even trained falcons. But they’re pretty costly, and can be unreliable.
The new method: In a recent paper in IEEE Transactions on Robotics, engineers from Caltech describe how they built on sheep herding algorithms to create software that could autonomously control a drone. The team learned how to gauge the proper distance between the drone and the animals by observing how multiple flocks of birds interact with each other.
The results: The team found the system to be effective in tests for flocks of dozens of birds with only one drone (see gif). The technology could help prevent accidents like Sullenberger’s “Miracle on the Hudson” landing.
The big new idea for making self-driving cars that can go anywhere
The mainstream approach to driverless cars is slow and difficult. These startups think going all-in on AI will get there faster.
Toronto wants to kill the smart city forever
The city wants to get right what Sidewalk Labs got so wrong.
Why sounds and smells are as vital to cities as the sights
The growing field of sensory urbanism is changing the way we assess neighborhoods and projects.
Marseille’s battle against the surveillance state
The boisterous, rebellious port city is trying to fight the growing ubiquity of policing cameras.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.