Microsoft teaches self-flying gliders how to wing it
HAWTHORNE, Nev. — As the glider turned and flew south, four men gave chase in a sport utility vehicle, rolling through the Nevada desert.
From the front seats, two of the men tracked the glider by sight. The other two followed the flight on their laptops, eyeballing data sent from the glider’s tiny computer and barking the figures into a walkie-talkie. In a Jeep up ahead, Ashish Kapoor listened as he, too, sped down the gravel road, eyes fixed on the white Styrofoam glider.
Soon, the glider took another turn. It gently circled an invisible column of rising hot air while climbing slowly. “It’s soaring,” Kapoor said as it spiraled higher and higher on a stream of warm air. “It found a thermal.”
Last week, in a desert valley 130 miles south of Reno, Kapoor and other Microsoft researchers tested two gliders designed to navigate the skies on their own. Algorithms learn from on-board sensors and predicted air patterns to plan a route, helping these gliders seek out thermals — columns of rising hot air — and use them to stay aloft.
The hope is that the autonomous aircraft can eventually fly for hours or even days while consuming little power, helping to, say, track weather patterns, monitor farm crops or even deliver the Internet to places where it is otherwise unavailable.
Led by Kapoor, an artificial intelligence researcher and a licensed pilot, the project was part of a growing effort to build aircraft, automobiles and other machines that can make decisions on their own when faced with uncertainty — an essential skill for any machine trying to navigate the world on its own.
Using similar methods, Google has built high-altitude Internet balloons that can stay aloft for months. Countless companies are designing cars that can drive on their own. And academics at schools like UC Berkeley are developing everything from household robots that can perform seemingly simple but surprisingly complex tasks like making a bed to surgical robots that can handle some procedures on their own.
Cars, planes and other robots can now recognize the objects around them with an accuracy that rivals human sight thanks to the recent rise of neural networks, a term for mathematical systems that can learn certain tasks by analyzing vast amounts of data.
But that only gets them so far. To navigate the world on their own, they must also mimic the way humans intuitively predict what will happen next and adjust their behavior accordingly. Projects like those at Microsoft, Google and Berkeley are reaching in that direction.
This kind of research has become increasingly important as Google and many companies try to build driverless cars. Mykel Kochenderfer, a Stanford professor of aeronautics and astronautics, said Microsoft’s project is a step toward self-driving vehicles that are nimble enough to handle all the unexpected behavior that human drivers, bicyclists and pedestrians bring to public roads.
It is also a way of pushing the boundaries of the mathematical techniques that control a machine in a relatively safe but still very real environment. “With a glider, you can test these algorithms with minimal risk to people and property,” Kochenderfer said.
Kapoor and his team relied on techniques that date back decades — something called Markov decision processes. Essentially, this is a way of identifying and responding to uncertainty.
The approach is like the one you take when looking for change in a backpack crammed with random stuff. If you just stick your hand in the bag and start rummaging around, you face enormous uncertainty. You do not know where to grab. But if, first, you remove the larger items like books and pencils that you know are not coins, the change falls to the bottom and the task gets easier. Microsoft’s algorithms do that in a mathematical sense, working to reduce the scope of the problem.
The team included Andrey Kolobov, who specializes in these methods. When he joined Microsoft’s research group four years ago, Kolobov fed these ideas into Windows and Bing. Back then, he was dealing with uncertainty in the digital world. Now, he’s applying them in the physical world. “The number of applications where these methods are used is growing,” Kolobov said.
In the Nevada desert, the team launched its two gliders with help from a remote control. Once airborne, the gliders — or sailplanes — were left to their own devices. They were forced to fly with help from the wind and other air patterns.
Through those onboard algorithms, the gliders could analyze what was happening and change directions as need be. They could learn from their environment, and although they could never be sure what would happen next, they could make educated guesses. Because it is dependent on phenomena it has no control over, the glider must reason and plan in advance, Kolobov said.
The gliders planned their paths to places that could provide lift, and then worked to ride those columns of rising air.
Still, the aircraft were far from perfect. Using a fiberglass glider with a 16-foot wingspan, the team hoped to set a record for autonomous flight time by a sailplane — more than five hours aloft. But in two days, that didn’t happen.