Trending: One year later, Microsoft AI and Research grows to 8k people in massive bet on artificial intelligence

Microsoft AI sailplane
Microsoft’s autonomous glider soars through the air above Hawthorne, Nev. Once airborne, the glider uses artificial intelligence to find and rely on thermals, or columns of air that rise due to heat, to stay aloft. (Microsoft Photo / John Brecher)

Paying attention to the “rise of the machines” increasingly means scanning the skies for things other than conventional aircraft or birds. But what if the line between the two begins to blur and autonomous planes can somehow be taught to mimic nature?

That’s the hope of researchers from Microsoft who are using artificial intelligence to keep a sailplane aloft without the help of a motor. A new report on the Redmond, Wash.-based tech giant’s website details the efforts of scientists launching test flights in a Nevada desert.

The researchers have found that through a complex set of AI algorithms, they can get their 16 1/2-foot, 12 1/2-pound aircraft to soar much like a hawk would, by identifying things like air temperature and wind direction to locate thermals — invisible columns of air that rise due to heat.

“Birds do this seamlessly, and all they’re doing is harnessing nature. And they do it with a peanut-sized brain,” Ashish Kapoor, a principal researcher at Microsoft, said in the report.

Kapoor said it’s probably one of the few AI systems operating in the real world that’s not only making predictions but also taking action based on those predictions. He said the planes could eventually be used for such things as monitoring crops in rural areas or providing mobile Internet service in hard-to-reach places.

Beyond those practical tasks, Andrey Kolobov, the Microsoft researcher in charge of the project’s research and engineering efforts, said the sailplane is charting a course for how intelligent learning itself will evolve over the coming years, calling the project a testbed for intelligent technologies. It’s becoming increasingly important for systems of all kinds to make complex decisions based on a number of variables without making costly or dangerous mistakes.

Read more about what Microsoft is learning this summer in the desert via the story from the company’s News Center.

Subscribe to GeekWire's Space & Science weekly newsletter

Comments

Job Listings on GeekWork

Find more jobs on GeekWork. Employers, post a job here.