Wings That Leave the Water

A Half-Pound Machine With a Bird's Ambition

Picture a puffin leaving the sea. It has just been swimming after prey. Now it has to punch back into air that is, roughly, a thousand times less dense than the water it is leaving. About a hundred bird species manage some version of this double life — loons, gulls, petrels, kingfishers among them. Same wings. Two utterly different worlds.

For a long time, no mobile robot had joined them. A team led by Raphael Zufferey, assistant professor of mechanical engineering at the Massachusetts Institute of Technology, has now built one that does. With colleagues at EPFL in Lausanne, Switzerland, and Northwest Indian College in Bellingham, Washington, they designed what they call a flapping-wing aerial-aquatic vehicle — FAAV, a small winged robot meant to travel in both air and water. It weighs less than 300 grams, about half a pound.

In a tank, and later on Lake Geneva, it swam up from about half a meter down, broke the surface, and kept flying. The results, set to appear in the journal Science, are as much a question about birds as they are a machine. How do diving birds retune a wingbeat when the stuff around them suddenly gets a thousand times thicker? And could a robot that small someday carry both the answer and a water sample home?

Borrowing a Tempo From Real Divers

Zufferey runs MIT's AURA Lab, where the robots stay small on purpose, so they can slip into oceans and waterways without making a fuss. This one looks, loosely, like a bird: a central body, two flexible wings, and a steerable tail. Inside sit a battery and a waterproof electric motor. The motor turns a crankshaft — the same idea as the rod that pushes a bicycle pedal in a circle — and that crank pumps the wings up and down at a chosen beat.

Before any of that hardware, the team went looking in the scientific literature. Puffins, petrels, kingfishers, and other divers had already left clues. Smaller birds flap about ten times a second in air and about four times a second in water. Larger birds beat a little slower, because their wingspans are wider. Puffins can swim at about 3 meters per second, a brisk human walking pace, entirely underwater.

"You have to do some adaptation to make that transition work," Zufferey says. "But there's a solution that exists in nature." Birds, he notes, already do pretty amazing things. No one had tried the same leap in a mobile robotic system.

The robot's wings are thin membranes coated with hydrophobic nanoparticles — tiny water-shunning specks that help the surface shed droplets instead of hauling a wet coat into the air. The tail is motorized, so it can tilt and point the body up or down. Wings and tail can be swapped. The team built three spans and tried them all: 60 centimeters, 80 centimeters, and 100 centimeters across.

The Steep Little Doorway Out

They set the robot about half a meter underwater and changed three things at a time: wing size, how fast the wings beat, and the angle of the tail. The combination that let it reliably swim, break the surface, and fly used the medium wings.

Flexibility was the quiet requirement. The wings have to be soft enough that their stroke stays modest in water — a wide, forceful flap down there would be fighting density itself — and firm enough to hold the robot up once it is in air. At about five flaps per second (engineers write that as 5 hertz; it just means five beats), the robot swam at nearly 1 meter per second. In air, a similar beat carried it at about 6 meters per second. Those speeds and rhythms sit close to what real diving birds do.

The leap wanted a steep nose-up pitch of 70 degrees. That angle keeps the wingtips clear of the surface as they beat upward. Any steeper, and the robot tips back into the water. It is a narrow doorway, carefully mapped — not a flaw in the idea, but a precise instruction for the next flight.

Then came the result that makes the machine feel like more than a clever copy. Puffins and ducks usually paddle their feet at the surface while wings flap and tails pitch. This robot has no feet. It still left the water.

"If you look at birds, most birds need to paddle at the surface to take off," Zufferey says. "And the question was, do we need the same for robots? And it turns out we don't."

Errands the Ocean Has Been Waiting For

The dream he describes is practical, and a little tender. Oceanographers, marine biologists, and people who live along the coast could launch the robot from a boat or from shore. It would fly toward an iceberg, a port facility, or a pod of whales, dive to take a measurement or collect a sample, and fly the data back — at a fraction of the cost of sending a traditional vessel. Then it could go out again.

That future is still ahead of the hardware, and the team is treating the gap as the next piece of work rather than a disappointment. They are redesigning the wings so they can turn, not only flap up and down. They also plan to test choppy water and wind — the messy air and water a calm lake does not fully stand in for. Those trials are how a promising step becomes a tool someone can trust offshore.

One of the stubborn problems in ocean science is getting measurements often, and in many places at once. A robot that can leave, dive, return, and leave again might someday do that not once a week, but every hour. "You could send this out not just every week, but every hour," Zufferey says. "It could fly out at high speeds, dive in fly back, deliver its data, and go back out, multiple times."

A half-pound set of wings, beating at something like a bird's tempo, has already shown that the crossing from water to air does not have to wait for feet. What it does next — in wind, in chop, over a real stretch of sea — is the part still opening.

"Our dream vision is for oceanographers, marine biologists, and members of coastal communities to launch this robot from a boat, or from shore, and it would fly close to the area of interest." - Raphael Zufferey