Airplane wings trade off some of their efficiency for stability. Their shape is fixed and rigid except for the flaps and slats that pilots can tune during takeoffs and landings to attain the desired level of lift and drag. Birds approach flying a bit differently: They take off, glide, turn, and land more efficiently and nimbly than airplanes by changing the shape of their flexible wings to various configurations (see Physics Today, June 2007, page 28).
The feathers that make such morphing possible are made of a few basic parts. Each feather consists of a central shaft that’s lined with branching barbs. The many hook-like barbules on each barb easily fasten together to form an individual feather with a pliable and gapless surface. Based on several detailed observational studies, biologists have assumed for decades that simple friction causes feathers to cling together during flight.1
Scientists could use computational fluid dynamics simulations to test, in principle, whether frictional forces are strong enough to keep feathers together. But the research is complicated by the morphable wings, which deform as they interact with air. More recently, engineers have begun applying dynamical systems analysis to better understand how individual structures in bird anatomy, including the neuromuscular and sensory systems, integrate to control wing shape and flight.2
A group of biologists and engineers that recently teamed up has now uncovered the biophysical mechanism that makes morphing feasible. Graduate student Laura Matloff, her adviser David Lentink (both at Stanford University), and their colleagues determined that adjacent feathers on the wings of common rock pigeons are held together by microstructures similar to barbules.3 Field tests with a biohybrid flying robot, shown in figure 1, confirm that the microstructures fasten together in one direction and unlock in the other to prevent wings from getting bent out of shape in flight.
This biohybrid robot combines mechanical parts with pigeon feathers. During real-world flight tests with high and low turbulence, PigeonBot controlled the position of its 40 feathers using four wing-mounted motors. The feathers’ fasteners prevented any gaps in the wings from forming that would have otherwise limited the robot’s mobility. (Lentink Lab/Stanford University.)
This biohybrid robot combines mechanical parts with pigeon feathers. During real-world flight tests with high and low turbulence, PigeonBot controlled the position of its 40 feathers using four wing-mounted motors. The feathers’ fasteners prevented any gaps in the wings from forming that would have otherwise limited the robot’s mobility. (Lentink Lab/Stanford University.)
Prime mover
The researchers first measured the skeletal kinematics of pigeon wings and their relative flight positions using cameras and microcomputed tomography scanning. They couldn’t observe each configuration of the bird’s fingers, arms, and wrist bones in an active, living bird, so three pigeon cadavers were used for the experiment. The wings of each cadaver were positioned in various glide poses to collect the necessary data.
Biologists already know that the distribution of feather positions arises from bone movements that are mediated by a series of muscles and elastic ligaments. The scans and images of the wings supported that view, but the group found another correspondence between feathers and bones: As the pigeon’s wings morph for flight by extending the angle of the wrist bone, the angle of the feathers relative to the forearm bone increases linearly.
The transfer functions—that is, the translation of a bird’s wrist bone angle to each of the individual feather angles—can be generally described as an underactuated system. Computer scientists, mathematicians, and roboticists use the term for mechanical systems that contain fewer actuators than degrees of freedom. In the case of birds, the lower arm, hand, and a finger bone actuate 40 flight feathers to position them collectively rather than individually. Lentink says, “It’s an ingenious system that greatly simplifies feather position control.”
Show of force
When Matloff and other graduate students presented the research at the annual meeting of the Society for Integrative and Comparative Biology in 2016, researcher Teresa Feo with the Smithsonian National Museum of Natural History suggested that they measure the friction between the feathers. “Laura Matloff and I decided to simply slide some pigeon flight feathers over each other in the lab during an afternoon and expected nothing beyond a little friction,” says Lentink. “Instead we suddenly felt the feathers lock together with a force that was something else.”
The force was not caused by friction alone: The calculated coefficient of friction for even the lowest measured normal force was far greater than the theoretical limit based on the feathers’ material properties. Some other mechanism was clearly causing, or at least contributing to, their resistance to separation.
To look at the feathers in more detail, Feo and Lentink’s graduate student Lindsie Jeffries used x-ray and scanning electron microscopy. They found microstructures that fasten together and apply a directional, opposing force, as shown in figure 2. The structures interlock when overlapping feathers—for example, the P9–P10 and P5–P6 pairs—are extended for flight. Because the micro-fasteners are directional, a pigeon can easily change the shape of its wings by unlocking them so that the feathers overlap more.
The opposing force FO between adjacent feathers of rock pigeons rapidly increases as the feathers slide apart. (Shaded regions show the standard deviation.) The magnitude varies between feather pairs P9–P10 (a) and P5–P6 (b). The normal force FN for both pairs is too small for frictional forces to keep the feathers together. (Adapted from ref. 3.)
The opposing force FO between adjacent feathers of rock pigeons rapidly increases as the feathers slide apart. (Shaded regions show the standard deviation.) The magnitude varies between feather pairs P9–P10 (a) and P5–P6 (b). The normal force FN for both pairs is too small for frictional forces to keep the feathers together. (Adapted from ref. 3.)
The discovery of directional fasteners prompted Matloff and Feo to look for the mechanism in 15 other bird species. They found that 12 of them use the fasteners to control wing shape and prevent feathers from separating in flight. Barn owls, barred owls, and chuck-will’s-widows lack them. The evolutionary loss does have a benefit: Barn owls can fly some 40 dB more quietly than pigeons, which makes them adept nocturnal hunters.
Learning to fly
The team applied the underactuation results and the fastener discovery to build PigeonBot—a biohybrid robot.4 Rather than attempt to mimic bird feathers with synthetic materials of carbon and glass fiber, the researchers used the real thing. The prototype employed four motors mounted on the wings to control the movement of synthetic ligaments and the position of 40 feathers, which lock together when extended.
During real-world test flights under both calm and turbulent conditions, PigeonBot’s wings morphed and extended at frequencies similar to those observed in pigeons. With fully extended wings, PigeonBot glided at realistic speeds and angles of attack. The directional fasteners easily unlocked so that PigeonBot could, for example, flex its wings to a tucked position for turns, like real birds do. In wind-tunnel experiments without the underactuated ligaments and fasteners, the angles of the feathers were purposefully extended and gaps formed between the feathers, which would have made the robot fly unreliably in the turbulent atmosphere.
“I think this research by the Lentink group is an excellent example of a general trend that’s starting to happen,” says Simon Sponberg, a biophysicist at Georgia Tech. “This fastener mechanism is showing the deeper physics understanding underneath the interesting biological structure that we’ve appreciated.” The microscopic dynamical properties of the feathers influence the macroscopic shape and kinematics of wings. Sponberg says, “I’m really excited about this idea that you can take interesting observations done in biology and connect the emergent dynamics with the functions that make biological systems so cool.” The insights from the new results may help guide designers toward improving aircraft, especially low-speed drones.
