Author Topic: Lower cost / high safety (around users etc) artificial muscles / actuators  (Read 1595 times)

0 Members and 1 Guest are viewing this topic.

Offline jasonheadTopic starter

  • Jr. Member
  • **
  • Posts: 10
  • Helpful? 0
So I've been exploring twisted nylon muscles for a bit, as it seems they could have some potential for cheaper robotics.

But they are quite a bit of work to build, and I still have a ways to go to do useful things with them.
http://www.livescience.com/43536-yarn-muscles-100x-stronger-human-muscles.html
One challenge is that an active cooling system will need to eventually be put in place.

One thing I have been wondering is, For the goal of a dextrous, high precision, affordable robotic hand system if it is the best way to go at this point?
I'd like to get started actually building a system, rather than be in long-term muscle manufacturing mode. And I'd like to stay away from using traditional motors for direct actuation.

Air muscles have seen some success, and that seems it might be the next likely candidate.
http://www.imagesco.com/articles/airmuscle/AirMuscleDescription02.html
I like the idea of air better than hydraulics overall - less cost and potential mess.
Challenges could be the small high pressure hoses and valves, with the compressor tank / compressor adding cost as well.

Twisted lines looks interesting, but it seems inherently limited by the repeated movement that will contribute to eventual material failure.
http://www.hizook.com/blog/2015/01/13/twisted-string-actuators-surprisingly-simple-cheap-and-high-gear-ratio

A linear electro-magnetic motor seems it could be interesting, a design like, with some gearing for precision control seems like it could have potential. I can think of a design that might work.

What do you all think?
The best combination of lower cost, and precision control potential.

Offline artbyrobot1

  • Jr. Member
  • **
  • Posts: 31
  • Helpful? 2
All of those options are too expensive or too "lab only" for the average person to use.  I'm making a fully actuated, realistic humanoid hand using servo motors mounted in the arm to control the wrist and fingers using a sort of pulley system like bike brakes use.  I'm planning to have something like 20 motors to control the various finger directions and using human bone shapes and artificial ligaments made of spandex material which is sewn over and into the bones.

Offline jasonheadTopic starter

  • Jr. Member
  • **
  • Posts: 10
  • Helpful? 0
I pretty much came to the same conclusion after talking to someone at the DARPA Robotics Challenge Finals.  I want to eventually actually build a production system, so the artificial muscle approach is best left to when there are significant funds and/or a full venture.
Like this :)
http://www.disneyresearch.com/publication/high-performance-robotic-muscles/

What about basic spandex ligaments, and then a thin spandex glove over the whole hand?

Offline artbyrobot1

  • Jr. Member
  • **
  • Posts: 31
  • Helpful? 2
Well that could work, but for my project I'm going for realistic skin so I'll have a silicone glove painted to look like skin that goes over the spandex ligaments.  It's gonna be SWEET!

 


Get Your Ad Here