Author Topic: Muscle Wire Control help!!!  (Read 2492 times)

0 Members and 1 Guest are viewing this topic.

Offline xamirxTopic starter

  • Jr. Member
  • **
  • Posts: 8
  • Helpful? 0
Muscle Wire Control help!!!
« on: June 20, 2008, 07:31:47 AM »
Hi everyone,

My final thesis is to produce a working binocular motion system for two eyeballs. My supervisor has expressed that they look as close to human eyes and motion as possible. My first thoughts were Flexinol muscle wires, as i thought they would do the trick being the closest thing to an actual 'muscle'.

But... after reading up further i found that these devices are fiercely non-linear due to the large hysteresis curve. This makes accurate position control very hard. Does anyone know of a method of control to do position control? Or can anyone steer me in the direction of some information on a simple position control for a sma (shape memory alloy)?

Another option would just be to buy a servo linear actuator but i can't seem to find any information that produces readymade sma servos.

thanks

amir

Offline AndrewM

  • Robot Overlord
  • ****
  • Posts: 255
  • Helpful? 0
Re: Muscle Wire Control help!!!
« Reply #1 on: June 20, 2008, 09:38:08 AM »
Any of the memory alloys are going to be too imprecise to accomplish what you are looking at.  Plus they don't return to shape quickly enough to allow the eyes to move back in the opposite direction.  You can use standard servos with a string or wire on a small pulley system to accomplish this pretty easily.
blog: www.iamwhen.com
Saving the world from humanity one robot at a time.

Offline benji

  • Supreme Robot
  • *****
  • Posts: 832
  • Helpful? 0
Re: Muscle Wire Control help!!!
« Reply #2 on: June 20, 2008, 09:48:01 AM »
true,, servos arnt linear when it comes to precise motion, i suffered with it then did use a stepper with 1.8 degrees as one step
anyways you can do many expirements with the servo and get a very close equation
with some math you can get very close to linearity
or you should search for expensive servos
good ol' BeNNy

Offline Asellith

  • Contest Winner
  • Supreme Robot
  • ****
  • Posts: 648
  • Helpful? 9
  • "I'm a leaf on the wind. Watch how I soar"
Re: Muscle Wire Control help!!!
« Reply #3 on: June 20, 2008, 12:08:22 PM »
Just use pulleys to gear down the servos. Make a system that would allow you to move like 300 degrees on the servo and make that the full range of you eyes so just shooting from the hip here but lets say the eye moves a may of 25 degrees. So you error rate in your servos will be divided by 12. So even if the servo moves 5 degrees your eye only moves like .41 degrees. That allows you to cancel out the errors from non-linearity and bad feedback. The pulley system may be rather complex and might introduce some error on its own. Add some springs to give you tension in any position except centered. It might work rather smoothly then as long as you don't put to much tension on it.
Jonathan Bowen
CorSec Engineering
www.corseceng.com

Offline Admin

  • Administrator
  • Supreme Robot
  • *****
  • Posts: 11,658
  • Helpful? 169
    • Society of Robots
Re: Muscle Wire Control help!!!
« Reply #4 on: June 21, 2008, 05:27:25 PM »
Use regular servos, and tie strings to them attached to the 'eye' for muscle-like action.

Or
http://www.firgelli.com/

Offline xamirxTopic starter

  • Jr. Member
  • **
  • Posts: 8
  • Helpful? 0
Re: Muscle Wire Control help!!!
« Reply #5 on: June 22, 2008, 02:58:01 AM »
Thank you so much for the help guys. I think I'm going to try the linear actuator method or if the budget is too tight, the string and servomotor sounds like viable option.

thanks again!

amir

Offline izua

  • Supreme Robot
  • *****
  • Posts: 682
  • Helpful? 0
Re: Muscle Wire Control help!!!
« Reply #6 on: June 22, 2008, 03:13:48 AM »
wow, you got some killer funding to work with linis.
Check out my homepage for in depth tutorials on microcontrollers and electronics.

 


Get Your Ad Here