MIT closes in on bionic speed

[C1ay]

Robots, both large and micro, can potentially go wherever it's too hot, cold, dangerous, small or remote for people to perform any number of important tasks, from repairing leaking water mains to stitching blood vessels together.

 

lefthttp://hypography.com/gallery/files/9/9/8/soliton_thumb.jpg[/img]Now MIT researchers, led by Professor Sidney Yip, have proposed a new theory that might eliminate one obstacle to those goals -- the limited speed and control of the "artificial muscles" that perform such tasks. Currently, robotic muscles move 100 times slower than ours. But engineers using the Yip lab's new theory could boost those speeds -- making robotic muscles 1,000 times faster than human muscles -- with virtually no extra energy demands and the added bonus of a simpler design. This study appears in the Nov. 4 issue of the journal Physical Review Letters.

 

In this case, a robotic muscle refers to a device that can be activated to perform a task, like a sprinkler activated by pulling a fire alarm lever, explains Yip, a professor of nuclear engineering and materials science and engineering.

 

In the past few years, engineers have made the artificial muscles that actuate, or drive, robotic devices from conjugated polymers. "Conjugated polymers are also called conducting polymers because they can carry an electric current, just like a metal wire," says Xi Lin, a postdoctoral associate in Yip's lab. (Conventional polymers like rubber and plastic are insulators and do not conduct electricity.)

 

Conjugated polymers can actuate on command if charges can be sent to specific locations in the polymer chain in the form of "solitons" (charge density waves). A soliton, short for solitary wave, is "like an ocean wave that can travel long distances without breaking up," Yip adds. (See figures.) Solitons are highly mobile charge carriers that exist because of the special nature (the one-dimensional chain character) of the polymer.

 

Scientists already knew that solitons enabled the conducting polymers to conduct electricity. Lin's work attempts to explain how these materials can activate devices. This study is useful because until now, scientists, hampered by not knowing the mechanism, have been making conducting polymers in a roundabout way, by bathing (doping) the materials with ions that expand the volume of the polymer. That expansion was thought to give the polymers their strength, but it also makes them heavy and slow.

 

Lin discovered that adding the ions is unnecessary, because theoretically, shining a light of a particular frequency on the conducting polymer can activate the soliton. Without the extra weight of the added ions, the polymers could bend and flex much more quickly. And that rapid-fire motion gives rise to the high-speed actuation, that is, the ability to activate a device.

 

To arrive at these conclusions, Lin worked from fundamental principles to understand the physical mechanisms governing conjugated polymers, rather than using experimental data to develop hypotheses about how they worked. He started with Schrödinger's equation, a hallmark of quantum mechanics that describes how a single electron behaves (its wave function). But solving the problem of how a long chain of electrons behaves was another matter, requiring long and complex analyses.

 

This research was funded by Honda R&D Co. and the Defense Advanced Research Projects Agency/Office of Naval Research. Yip and Lin's collaborators on the work are Professor Ju Li at Ohio State University and Professor Elisabeth Smela at the University of Maryland.

 

Source: MIT

[Tormod]

This really sounds like science fiction coming true. I wonder if Honda will become the US Robotics from Isaac Asimov's novels. :friday:

[nkt]

What a badly worded press release! And it was from MIT!

 

Not much science in it, either. Have they actually got a single fibre working yet? Or is it still totally a theoretical speed boost?

 

1000 times faster would be rather dangerous, I think. The forces involved in even a rapid contraction of a human leg can easily break a bone that is as strong as carbon fibre. At 1000 times faster, moving a dumbell weighing 50lbs will exert massive forces on everything involved.

[akahenaton]

What a badly worded press release! And it was from MIT!

 

Not much science in it, either. Have they actually got a single fibre working yet? Or is it still totally a theoretical speed boost?

 

1000 times faster would be rather dangerous, I think. The forces involved in even a rapid contraction of a human leg can easily break a bone that is as strong as carbon fibre. At 1000 times faster, moving a dumbell weighing 50lbs will exert massive forces on everything involved.

 

 

i agree with the bone breaking, and the 1000 times faster acceleration than human muscle. But hey they do have their uses.

[Bio-Hazard]

REMEMBER!

 

"...machines dehumanize.." :phones:

[nkt]

This is reported in this week's New Scientist.

Robo-muscles deliver power from within

 

* 25 March 2006

* From New Scientist Print Edition. Subscribe and get 4 free issues.

* Zeeya Merali

 

"ONE day you could find yourself sitting in a bar next to a humanoid robot, who is taking a shot of vodka to give himself the energy to go to work." So says Ray Baughman, a nanotechnologist at the University of Texas at Dallas, who has developed self-powered artificial muscles that could ultimately be used in robotic limbs and prosthetics.

 

Baughman and his colleagues have designed two types of artificial muscle by adapting fuel cells so that the power source is part of the muscle itself, rather than coming from a separate battery.

 

The first is made from a nickel-titanium "shape-memory" wire coated in a platinum catalyst. A mixture of oxygen and either methanol or hydrogen is passed over the platinum coating, which catalyses a reaction between the gases. This releases heat, which warms the wire and makes it contract. When the flow of fuel stops, the wire expands back to its original length. The wire muscle exerts 100 times the force of a natural muscle of the same size, Baughman says.

 

Methanol is a particularly useful fuel because it has such a huge store of energy, says Siegmar Roth, an artificial muscle specialist at the Max Planck Institute for Solid State Research in Stuttgart, Germany. Methanol-powered fuel cells have the potential to deliver 30 times the energy of a conventional lithium-ion battery of the same size. This would mean the muscles could generate sufficient power to move a prosthetic muscle without adding unnecessary weight.

 

The team is now trying to find practical ways to supply the muscle with fuel. One possibility, Baughman says, is to use a valve mechanism controlled by very slight movements that could be used by people with limited finger or arm mobility. Another challenge is to find ways to prevent the muscle from overheating as it contracts, he says.

 

The researchers' second artificial muscle is made from sheets of carbon nanotubes coated in a catalyst. It is not yet as powerful as the wire muscle, but could potentially overtake it, Baughman says. The nanotube sheet acts as a fuel cell electrode, and as the methanol or hydrogen reacts with oxygen above its surface, charge is transferred to the nanotubes, causing the sheet to expand. Neutralising the charge should make the sheet return to its original size, but the team has not yet succeeded in doing this. They are now investigating how this could be achieved.

 

The nanotube muscle can also act as a super-capacitor, storing electrical energy for later use, Baughman says.

 

From issue 2544 of New Scientist magazine, 25 March 2006, page 30