Leif Ristroph wanted to build the “simplest possible” flying machine. The applied mathematician at New York University glued together several tubes of carbon fiber to build this: a sphere with four wings attached to it that propels it as a jellyfish swims.
The flyer is only about eight centimeters in diameter — small enough to fit in the palm of your hand — and its mass is only about two grams, the equivalent of two paper clips.
Half of the mass is the motor, a commercially available component about the size of the vibrator in a phone. There’s no battery on board. For the time being, the flying jellyfish is tethered to a power cord.
What’s most remarkable about the device, though, is that even though it has absolutely no circuitry and no sensors, it manages to keep itself upright in the air. That’s apparently a physical property of the arrangement of the wings that Ristroph says he can’t fully explain, but he hopes it will allow his prototype to evolve into a new generation of very small and inexpensive drones.
Researchers have built slightly larger machines with impressive stability and control. Raffaello D’Andrea and his team at the Swiss Federal Institute of Technology in Zurich are well known for their “robotic athletes,” which have four helicopter rotors. The machines can catch a ball and perform other autonomous feats of balance and coordination.
The problem is that conventional propellers are ineffective at small scales, Ristroph explains. Other groups have tried to mimic birds and insects, designing robots with wings that flap. The Harvard Microrobotics Laboratory has built a device even tinier than Ristroph’s that resembles a three-legged mosquito.
Ristroph, though, wanted to try something different. “My initial thinking was basically, ‘Okay, insects beat their wings like this. Birds beat their wings like that. Everyone else is building their robots to copy that,’ ” he said. “Let me think of new ways to fly.”
He and his colleague, Stephen Childress, have considered a number of outlandish designs, including a machine with a single, conical wing that Ristroph believes could theoretically keep a machine aloft.
The flying jellyfish was particularly exciting.
The device would right itself in response to disturbances. Flying machines modeled on insects typically require a computer on board or fins of some kind to maintain their stability, but Ristroph wanted to do away with all of those systems entirely. Doing so would make the machine lighter, which in principle would allow for an even smaller motor and a cheap, miniature drone.
Of course, any practically useful version of the jellyfish would have to carry a small battery and a navigation system — right now, all the navigation controls are manual — but Ristroph is optimistic those problems could be solved.
A more difficult problem is explaining the machine’s stability in the air. It will wobble and flit back and forth in flight, like a moth, but never stalls. Each deviation is compensated for somehow, and the jellyfish stays aloft.
Ristroph wrote a list of formulas to account for the device’s behavior, but he didn’t find them persuasive.
“The assumptions I make in the model don’t seem to be justified,” he said. It’s a concession you don’t often hear from researchers.
He envisions flying jellyfish that cost 50 cents each. They could be deployed to monitor air quality, or they could just be disposable toys. “If someone steps on it, it’s fine,” he said.
He isn’t particularly excited about the possibility of a version of his jellyfish being used by the military, and he isn’t comfortable with the Obama administration’s use of drones. But, he said, “Having good drones could, overall, save lives also. I know it’s a big, thorny issue.”
Ristroph demonstrated the device at a conference Sunday, and he’s hoping other researchers will help him figure out why it works.