Small air muscle

I have been looking for alternatives for servo motors, and I stumbled apon air muscles. It seems to me air muscles are very promising

  1. they can do a significant amount of work compared to their size
  2. they are virtually indestructable
  3. they move smoothly in a natural manner
  4. they are cheap and easy to build yourself

There are probably drawbacks too, but I think there is not much you can say about it before someone has built a whole robot and see how it works out. There are two main challenges, and they are related:

  1. how to regulate air supply to each air muscle (electro-mechanical-pneumatical conversion)
  2. a feedback system with position/rotation sensors

I have done a simple test with different loads. I could do more detailed tests but the conclusion is simple: the relation between air pressure, load and displacement is a complex one. So a feedback system is really needed.

I would very much like to have some other people join this project. Before air muscles become a serious alternative a lot of work needs to be done. If you want to join, write a comment or send me a message.

https://www.youtube.com/watch?v=Goix8VuC9GU

Welcome to LMR.Finally I can

Welcome to LMR.

Finally I can see somebody is working on air mucles here on LMR. I had that idea a couple of months ago but abandoned it when I could not find a small and powerful enough pump to mount it on a mid size mobile robot.

Looking forward to further development of your :slight_smile:

Thanks

Thanks lumi,

I did not think much about the air source yet. I thought a 16 gram CO2 cartridge would be ok, they cost 2 dollar or so, and you even can make them refillable. Or an external compressor with a small on-board container? Operation time will be short, but recharging is a matter of seconds (both with a CO2 cartridge and an external compressor). Let me know if this works for you.

 

Well, CO2 cartridge might be

That should be a reply to your comment…

Well, CO2 cartridge might be the best solution but as the compressor…short operation time since the CO2 is limited and the compressor’s battery will run out fast too.

I guess that for mid size robots the air muscle is not applicable when the robot needs to be mobile. A stationary one like a robot arm would be ok since you could power the compressor by a power unit.

Interesting topic!

Hi Carloroosen,

This is an interesting topic, as it can open new developments for robotics.

As far as I know and saw in internet, air muscle actuator

s have good advantages, like the easy manufacturing, but also they have some issues like the need of a pressure generator that can be static like a CO2 cartridge mentionated or a pump, yes it’s like having a battery or power generator for electric actuators but with a series exchange of energy: from electricity or gasoline to air, and the air going to the muscle. As you may know this implies loose of energy. Electric actuators just need a direct conversion between battery to actuator so they are usually more efficient.

The presure pump must be also powerful enought for all the muscles you want to handle but small, robust and light enough to be inside/wearing by a robot… I did a little research in eBay and found some cheap ones that maybe can be useful for small or bigger robotic projects. Also you can make your own DIY pump… need some pneumatic valves… and check valves.

However I think it’s definitive an interesting topic to explore. Keep going and we’ll tall about it :wink:

Kind regards,

Francisco

Paintball

Just to throw it out there, you might want to look at paintball supplies. Basically, there are a bunch of air tanks, fittings, braided hoses etc. --And all off the shelf, cheap, easy to find and light weight aluminum.

what about this

Would this pump be in a usable range for a standalone robot, 12V, 12W ??

https://www.sparkfun.com/products/10398

At least it does give a nice pressure of 3 bar (according to documentation I found elsewhere). I ordered it, but since I live in The Netherlands it will take a few weeks. I’ll keep you updated )

 

what about this

Would this pump be in a usable range for a standalone robot, 12V, 12W ??

https://www.sparkfun.com/products/10398

At least it does give a nice pressure of 3 bar (according to documentation I found elsewhere). I ordered it, but since I live in The Netherlands it will take a few weeks. I’ll keep you updated )