Butterflies offer lessons for robots

By Kimberly Patch, Technology Research News

It turns out that butterflies' fluttering is neither random nor clumsy.

Researchers from Oxford University in England have devised a method of studying the way butterflies fly, and their initial results show that the insects have many more tricks of flight than they get credit for.

The researchers trained red admiral butterflies to fly between artificial flowers in a wind tunnel, and recorded the way air flowed around their wings using smoke and high-resolution cameras. The work provides fodder for researchers working on insect-sized flying robots.

Previous studies have revealed a couple of secrets of insect flight. Researchers from the University of Cambridge in England working with tethered hawkmoths and flapping robots showed that wing movement creates columns of spinning air, or vortices, above the leading edges of wings, which provides the lift needed to fly. A vortex above a wing can create as much as a twofold increase in lift.

And researchers from the University of California at Berkeley working with robotic fruit fly wings discovered that fruit flies capture turbulence from their own wakes to increase lift.

These results put insects in two camps - large insects that produce vortices above the leading edges of their wings to create lift, and small insects that instead hover or fly slowly enough to capture lift as their wings pass back through the wakes of disturbed air they leave behind.

The free-flight studies proved butterfly flight is much more complicated, according to Robert Srygley, a professor of behavioral ecology at Seoul National University in South Korea and a research associate at the University of Oxford.

Free-flying butterflies "use all of the known mechanisms to enhance lift -- wake capture, leading-edge vortex, clap and fling, and active and inactive upstrokes -- as well as two mechanisms that had not been postulated, the leading-edge vortex during the upstrokes and the double leading-edge vortex," said Srygley.

The research showed that butterflies create vortices and double vortices above the leading edges of their wings by varying the twist and speed of their strokes to make sudden changes in pitch.

They use vortices shed from the wing's trailing edge -- the wake -- recycling their own energy to further increase lift. They also use a clap-and-fling mechanism to produce opposite vortices on each wing, which also contributes lift. The red admirals do this by touching their wings briefly, then rapidly separating them. And they use a mix of active upstrokes, which generate lift, and inactive upstrokes, which do not.

The basic butterfly stroke is not smooth. Going into a downstroke each wing is up and back, with the wings' leading edges pointing foreword. As the wings go down and foreword they also continuously rotate, changing the wing angle. Just before the upstroke the butterfly quickly twists its wings so the leading edge points backwards. On the upstroke the wings go up and back and again continuously rotate; there's another quick rotation at the end of the upstroke to position the wings before the next downstroke.

The brief stops before each downstroke and upstroke, and accelerations and decelerations between strokes vary the airflow considerably.

The researchers' smoke patterns showed that butterflies often used different aerodynamic mechanisms in successive strokes.

The familiar, random-looking fluttering of butterflies is really due to the animals using a wide variety of aerodynamic mechanisms as they take off, maneuver, maintain steady flight, and land, said Srygley.

In general, the butterflies made more use of vortices during acceleration, Srygley said. "The leading-edge vortex is most pronounced when the red admiral butterflies are accelerating; when maintaining a steady speed [it] became less pronounced," he said.

This makes sense given the drag that must arise from altering the flow of air from across the top surface of the wing to form a vortex that moves towards the wing tip, he said. "Insects are probably trying to minimize the drag during steady foreword flight, and restrict use of the leading-edge vortex to periods of acceleration and maneuvers," he said.

It is interesting that the butterflies show various wing aerodynamics during different modes of flight, said Robert Michelson, a principal research engineer at the Georgia Tech Research Institute. And it's significant that the butterflies were not tethered, but allowed to fly freely.

The work runs counter to a study at Cambridge University in England that showed that the leading edge vortex varies in diameter as it moves out along the wing during the flap, said Michelson. "This study seems to counter the notion of diameter change and span-wide flow," he said. "A third validating study would be nice to help resolve who is right."

Michelson's research includes a small, flapping wing robot dubbed Entomopter. Flapping wing aerodynamics is not well understood, said Michelson. "This study adds to the rather meager body of knowledge," he said. Controlling insect wings "is physically complex, difficult to miniaturize, and... very power hungry," he added.

The study improves the understanding of aerodynamics at a scale where engineers have not yet built many systems, but where nature has a great diversity of designs, said Ron Fearing, a professor of electrical engineering at the University of California at Berkeley.

The results could be useful for robotic fliers weighing between 100 milligrams and 10 grams, he said. "If the aerodynamic efficiency were significantly higher than fruit fly kinematics, it would likely be worth using a more complicated wing drive mechanism," he said. The mechanical difficulty of using the more complicated butterfly kinematics and the precision of control required has to be evaluated, he added.

There is a lot left unknown about insect flight, said Srygley. "Just about every flight mechanism in the insect world remains unexplained," he said.

The basic theory for biological flight is based on propeller theory. This theory adapted to flapping wings does a reasonable job of explaining bird flight, said Srygley. "However, it does not explain all the lift required for an insect to fly."

The researchers next step is to explore the diversity of insect flight using the free flight study method, said Srygley. "We've just opened the door on free flight studies, and of course much remains to be discovered."

The research could find use in robotics within a decade, said Srygley. "I would expect that we will see flapping [robots] the size of butterflies or hawk moths with reasonable flight durations [and] distances in five to ten years," he said.

Flying robots could explore volcanic vents, assess stresses on bridges or skyscrapers, or other planets, said Srygley. "Hundreds of small robots could be lifted into space to probe planetary surfaces rather than lifting a single crawling robot," he said. As long as the planet to be explored has an atmosphere, more area could be covered using flying robots, he said.

Srygley's research colleague was Adrian L. R. Thomas of the University of Oxford. The work appeared in the December 12 issue of Nature. The research was funded by the British Biotechnology and Biological Science Research Council (BBSRC).

Timeline:   5-10 years
Funding:   Government
TRN Categories:  Applied Technology; Biology; MicroElectroMechanical Systems (MEMS); Robotics; Physics
Story Type:   News
Related Elements:  Technical paper, "Unconventional Lift-Generating Mechanisms in Free-Flying Butterflies," Nature, December 12, 2002.




Advertisements:



February 12/19, 2003

Page One

Teleporting goes distance

Scheme smooths parallel processing

Butterflies offer lessons for robots

Social networks sturdier than Net

Logic scheme gains power




News:

Research News Roundup
Research Watch blog

Features:
View from the High Ground Q&A
How It Works

RSS Feeds:
News  | Blog  | Books 



Ad links:
Buy an ad link

Advertisements:







Ad links: Clear History

Buy an ad link

 
Home     Archive     Resources    Feeds     Offline Publications     Glossary
TRN Finder     Research Dir.    Events Dir.      Researchers     Bookshelf
   Contribute      Under Development     T-shirts etc.     Classifieds
Forum    Comments    Feedback     About TRN


© Copyright Technology Research News, LLC 2000-2006. All rights reserved.