Logo: University of Southern California

On the Threshold of Miniature Flight

What will the next generation of small-scale flying machines look like, asks a team of aerospace engineers. Bats? Birds? Dragonflies?

March 22, 2006 —
With the steadiness of a surgeon, John McArthur, a graduate student in aerospace and mechanical engineering, positions a slightly curved plate about the length of a 12-inch ruler on a rod that protrudes from the force balance in USC’s Dryden wind tunnel.

Geoff Spedding, left, watches as John McArthur aims a laser at his newly machined wing and adjusts the airflow for a flight experiment in USC's Dryden wind tunnel.
The flow properties and the force balance calibrations in this tunnel, housed in the basement of the Rapp Research Building, have been carefully tested over three or four months.

Today the airflow in the tunnel will be seeded by a fog of one-micron-diameter smoke particles and illuminated by consecutive double pulses from a dual-head laser.  Each light pulse will be five nanoseconds in length and consecutive flashes will be separated by about 150 microseconds. The laser will flash at a rate of 10 Hertz, so each pulse pair will be regenerated every 1/10th of a second. Images of the airflow will be captured on a charge-coupled device (CCD) array camera and saved in real time.

McArthur’s faculty adviser, Geoff Spedding, a professor of aerospace and mechanical engineering, and his colleagues have developed custom software over the last 10 years to perform an analysis of this flow field with extreme accuracy.  After making the measurements, they’ll be able to describe not only the lift and drag forces of this small curved wing, but also the spatial gradients of the airflow in which it was flying, which are difficult quantities to estimate.  

No one has paid much attention to small-scale airfoil geometry — how the shapes of small flying things impact their flight capabilities — or to understanding the aerodynamics of winged flight on small scales, until quite recently, Spedding says, as he adjusts the cambered plate and aims the laser.

Spedding collaborates with biologists at Lund University, Sweden, who study wing flapping as live birds fly through a wind tunnel.

“The geometry and kinematics of bird flapping are complicated,” he says.  “The question then arises, must this complexity be mimicked or are there more simple fundamental designs of small-scale aerodynamics that can be applied to build the next generation of small, remotely piloted flying machines?”

Winged Flight
Spedding and a research group in USC’s Aerospace and Mechanical Engineering (AME) Department are busy pursuing those questions with data from their wind tunnel experiments.  AME graduate students machine the simplest possible wings — flat plates, curved plates, and classical airfoils — and then plot the wind tunnel measurements, which sometimes turn out to be quite counter-intuitive, on graphs that are scotch-taped to the laboratory door.  

Winged flight has always fascinated Spedding, who is a zoologist by training. After earning a Ph.D. in zoology from the University of Bristol, England, he specialized in animal aero- and- hydrodynamics.  With the recent upsurge of interest in small-scale aerodynamics, he has begun to investigate what affects the aerodynamic performance of these very simple objects, which fly at very low speeds.   

“Imagine a flying ruler, which is a simple flat plate,” he says, holding up a ruler he has retrieved from the top drawer of a desk.  “How well could this fly when attached to a suitable airframe?   How would we improve this design?”

It is remarkable that such questions qualify as topics of research, but Spedding says there are two reasons for it. “First, all of our textbooks on aerodynamics and on aircraft and helicopter flight have been developed for devices that are much larger and fly much faster.  These aerodynamic models and analytical methods are, arguably, among some of the crowning intellectual and practical achievements of the past century.  

McArthur, left, examines the curved edge of a plate that he machined and will test in a later wind tunnel experiment.

“Modern aircraft are efficient and powerful, and routinely carry people and armaments over long distances,” he continues.  “But few have paused to reflect on how a very small plane might work.  In fact, most of the serious work has stopped at the scale of competition sailplanes.”

Of Bats and Thrushes
He wends his way out of the wind tunnel and back through the maze of instrumentation filling the basement of the Rapp laboratory, then climbs a flight of stairs leading to his office one floor above the wind tunnel.  Stepping inside, he reaches for a small plastic bat with a wingspan of about 20 centimeters dangling on a string from the ceiling.  

When he isn’t collaborating with a group of biologists at Lund University, Sweden, using live birds (thrushes) and bats, Spedding relies on wind-up or battery-powered toys, such as this red-eyed Halloween bat, to inspire him.  He winds it up and gives it a gentle shove.  The bat begins to flap its wings and flash its red eyes as it circles high above his desk.  Gaining momentum, it lifts into a higher orbit.  

“Newton’s laws of motion in action,” Spedding grins.

How do Newton’s laws of motion work on much smaller scales?  Just the same, Spedding notes. But because the viscosity of the air can no longer be ignored, the flows are very complex.  And that is the second reason this research must be done.  Even simple textbook problems become complicated.  Standard aerodynamic models do not give researchers the answers.
    
“Paradoxically, it is far easier to predict, analyze, and model the flow around a Boeing 747 than it is to predict the flow around our simple ruler flying at some reasonable speed,” Spedding says. “And much of the existing data in the literature is controversial and inconsistent, with little apparent incentive to force the issues to resolution.”

Measuring the wingspan of a thrush.


Then what sort of small-scale flying machine should he build?   

“We could build just about anything,” Spedding says.  “Suppose we built a small flying machine that can flit through crowded spaces, hover silently in a precise position, making observations of a moving target, abruptly reverse direction in times of danger, and do all of this for more than an hour.  Then it would report back to base as directed with images and other sensory information, such as chemical and pressure readings, visibility, heat and radioactivity measurements.”

It’s not the technology that is holding engineers back from building that plane, he says.  It’s possible to build all kinds of small-scale electromechanical devices these days.  It’s the design — what it would look like.  

“It could look like a dragonfly.  A moth.  A bat.  A bird.  Or none of the above,” he says.  

Biological Locomotion
Assistant Professor Eva Kanso and Professor Tony Maxworthy, who holds the Smith International Professorial Chair in Mechanical Engineering, are members of Spedding’s AME research team addressing those and other problems in biological locomotion.

Tony Maxworthy


In 1979, Maxworthy, a member of the National Academy of Engineering, was the first scientist to realize and demonstrate that many simple wings in oscillatory flapping motion will generate strong swirling currents, or vortices, of fluid at the front edge of the wing, and that the forces associated with this strong rotational motion will be both beneficial and controllable.  In many cases, the presence of these complex, time-varying fluid motions can make the difference between flying or being grounded.

Since then, the leading edge vortex (or LEV) has become a staple of those seeking to understand the aerodynamics of both insect and bird flight.  Maxworthy predicts that it is “very likely that any successfully engineered device will have to have some similar means of generating and then controlling such fluid motions.”

Similar principles apply to swimming fish, he says.  Researchers are currently working on experimental modeling of the forces and flow fields generated by fin and tail motions.

Kanso, a mechanical engineer who earned her Ph.D. from the University of California, Berkeley, is the third member of the team.  Interested in the interaction of shape changes with wake dynamics, she is attempting to unlock the mathematical keys to underwater locomotion.   

“If we were to build a swimming robot, should it resemble a jellyfish? Or an eel?” she asks.  “How many degrees of freedom does it need to have to be able to achieve a desirable forward or steering motion?”

Eva Kanso explains how a segmented object might move and propel itself underwater.


Dynamical Systems Theory
She answers questions like that with simple, yet powerful, tools drawn from dynamical systems theory, geometric mechanics, and computation.  The reduced modeling approach has already been able to demonstrate some amazing things: for starters, that simple cyclic shape changes will occur in an articulated three-segmented swimming machine placed in water.  

Voilá!  It swims.  
 
This line of research enables novel engineering applications, such as the design of biologically inspired vehicles, both micro and macro, that can propel themselves by undulating their shapes.

“Investigating simple systems like this helps to unravel the basic principles of aquatic locomotion,” Kanso says.  “I am interested in quantifying these principles and building on them…learning to improve on what we’ve already built.”

The research has a ways to go, but it is a first of its kind. The data are providing new insights into biological locomotion and the future of small-scale flying and swimming machines.  But like all good mysteries, the future is full of unanswered questions.

“There’s still a lot to learn from bird flight before a reasonable airfoil design can be proposed,” Spedding says, “or before we start pasting bird feathers on our small flying machines.”   

But the day will come when this planet is populated by both animals and small autonomous vehicles whose forms and behaviors may — or may not — mimic the natural world. That day isn’t too far away.
 
  
--Diane Ainsworth/Photos by Irene Fertik