Has Parallax ever considered a "muscle wire" robot ?

Bean

Just wondering if anyone knows if Parallax has ever considered a "muscle wire" robot.
They are semi-popular, but the one's I've seen suffer from inferior construction, I think Parallax could make a superior robot.
Just a thought.
Bean.


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Check out· the "SX-Video Display Module"

www.sxvm.com
·

Jon Williams

We could, but they're not much fun ... muscle wire is slow and doesn't really have much "muscle"

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax
Dallas, TX· USA

Ken Gracey

Hey Bean,

Shape memory alloy (SMA) has other issues. Because of minimal contraction, you have to use it to actuate a lever with the point of inclination near the SMA actuator (not sure if I'm using the right terms here). Worse yet, it takes a lot of current so you'd have to carry a high-current power supply. In fact, the most recent release of the most popular SMA robot kit (the Stiquito) doesn't even carry its own power supply. You kind of need to "hold" the 2AAA battery pack over the robot and hope the pressure you're putting on the power supply wires doesn't make the little Stiquito veer off his path.

I reviewed the latest book, built the robot and Jon wrote a code example for it. Honestly, it's a great robot for $50 and it does work, but extending the concept to a larger, practical robot seems to be a challenge.

I'm sure the SMA professionals could achieve more robotic applications than I have seen. SMA has some impressive industrial applications, so it seems logical that it could be used in a robot. But that power supply problem. . . but most of us have experimented with SMA and nobody has been impressed enough to propose a robot.

Ken

The Dead Bug

I have thought about this as well after building my own "Stiquito". I think there'd be more potential in "muscles" made from neoprene tubing that can be inflated by pressure from a CO2 cartridge. If one could come up with a practical valve system that was tiny and could open and close under the pressure of CO2 and be operated by logic levels, strong light weight robotics could be possible. I have considered some sort of solenoud operated needle valve or something...but have not gone far with the thought.

Of course with the recent developments in polymer muscles, eap.jpl.nasa.gov/ if we wait another ten minutes, "muscle wire" will be passe' anyway.
Bruce

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Name: Bruce Clemens

Work:· Clemensb@otc.edu
Good Stuff on my Blog: http://theDeadBug.journalspace.com

SteveW

It might be that SMA would make a better gas valve than it would a direct actuator. You've still got the high power requirements and slowness to contend with, but the short travel and springiness might be convenient, in this application. Dunno...

Steve

Ray Iddings

I built on of the Stiquito "Robots", and I use that term liberally. Its is painfully slow and if you overheat the muclewire it loses its memory. I was very disapointed when I built one.

Its funny talking about robots, in general, The BEAM ones are fun because I guess they are what they are, I have an ER1 and it was fun for awhile but my programing skills arn't ready for it and the API access is kind of wonkee (It does play hide and seek with the kids though).

The BOE has been a lot of fun for a few reasons the learning curve isn't that steep (the Master Curve is probably another story).
the documentation is great and it isn't as intimidating as say my ER1, If one of my kids accedently breaks the BOE-Bot, I'll Be upset but not bankrupt like the ER1.

Robert Schwartz

The Dead Bug said...
I think there'd be more potential in "muscles" made from neoprene tubing that can be inflated by pressure from a CO2 cartridge. If one could come up with a practical valve system that was tiny and could open and close under the pressure of CO2 and be operated by logic levels, strong light weight robotics could be possible. I have considered some sort of solenoud operated needle valve or something...but have not gone far with the thought.

There are solonoid valves that can be used to power a air muscle, and there are a few very interesting robots that use these. With those however, power (CO2 levels) becomes a very great issue. High pressure air would be an overall better choice when you consider weight and ease of refilling. Unless you can put a portable air compressor on the robot, you would need to refill it fairly often. At least with battery powered 'bots you can slap on solar panels and let them charge themselves.

LoopyByteloose

Alegorically, 'muscle wire' was a near perfect advertizing gimic, but it really falls short of the dreams that it evokes. [noparse][[/noparse]it is very slow, requires a lot of current, needs to be cooled for reciprocal movement, and costs a bundle]

I would love to hear about its sucessful applications, so that I could have some idea of how it might be useful. All I can think of is some kind of munitions time delay.

And the other side of this problem is that we robotics hackers still are without the electronic analogue for biological muscles that everyone thinks muscle wire should be.

It seems that the best we can come up with is rotary motors that are geared down enough to allow armature to spin with enough momentum for smooth movement or a stepper motor for precise limited motions.

Neither really gives a bone and muscle kind of a setup and the motors can really make a robot top heavy.

Additionally, while hobby servos are easiest to plug and play with, PWM is not as speedy or efficient as H-bridges since you need to encode and decode the motion (the PWM was intended for R/C, NOT for robots).

Also, it doesn't matter if you use regular motors with gears, hobby servos, or steppers - the price of electro-mechanical devices has not gone down anywhere near as much as semiconductors. It some cases I think it is going up due to lack of continued surpluses as engineers understand that a few good standard sizes are all they need.

What I am trying to say is that we really need to dream a little more and not just bolt together the available parts and boards. (I know it is a lot of fun!)

Without the creation of new visions, we are just tinkering. Also, robotics is highly dependent on mechanics and we have to accept that mechanical engineering has a greater role to play in robotic design. Electronics alone won't get you very far.

I think that Thomas Edison would never have done it this way.

Hopefully I haven't gotten too far off the track.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
G. Herzog in Taiwan

SteveW

>I would love to hear about its sucessful applications, so that I could have some idea of how it might be useful. All I can think of is some kind of munitions time delay

A colleague here has used the metal (not in wire form) for undersea pipe jointing. Heat a collar, slip it over, and let it grab the pipes to be joined as it changes back.
Don't know if it's still used, it was a while ago.

Steve

LoopyByteloose

So, the metal is mostly useful as an assembly and fastner process [noparse][[/noparse]static use] and not a moving part [noparse][[/noparse]dynamic use].· It really seems appropriate for quick and secure assembly in hostle environments [noparse][[/noparse]maybe it has a future in space station construction, emergency arctic repair, or emergency nautical repair].

The underwater environment has its own electrical limitations due to the conductivity of saltwater.· If a nicked extension cord falls off a dock while passing current, it only takes a minute or two for the salt water to cause it to fail.· Of course, fresh water is much less hostile.

Still, does anyone have a sucessful dynamic use.· I think I saw a claim that part of the Mars' Rover used muscle wire.· This seems a bit odd because of the high power consumption requirement and the slowness of response [noparse][[/noparse]maybe Mars is so· cold that the response is somewhat enhanced].

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
G. Herzog in Taiwan