World Beyblade Organization by Fighting Spirits Inc. Off-Topic Forums Your Creations

Design & Technology Project - A robotic hand! (Survey inside, please help!

Sep. 02, 2013  1:27 PM
 
[Survey in Research Stage]

Hey everybody, Manicben here!
I usually don't do this stuff, but for my last course before university I have to do a tech project. And this year I chose to do a robotic hand using one or two smart materials, like shape memory alloys. Bit silly considering how hard it is, but despite that I shall try my best.
I shall keep the thread updated with photos and such to show ya'll the progress.


RESEARCH STAGE

Currently being worked on.

I would appreciate the help from everybody with suggestions and such.
Right now a way to help out is to fill out a web survey I have prepared.
Here's the link - http://www.surveymonkey.com/s/8VDHT62
Oh yeah, serious answers too please. Anyone can fill it out!
Using these data sets, I shall create and refine the design based on suggestions or even specs/features requested, as long as they are possible.

The situation is that the Science Museum in London requires a new exhibit related to technology. It has to be interactive, somewhat like the stuff at the Launchpad there for those who have been. They have requested that a robotic hand is made to demonstrate the concept and entertain people of different age groups.

Currently I plan on using SMA's along with an Arduino based Sainsmart UNO R3 microcontroller to program the hand to do things, such as movement and interactivity could involve playing Rock-Paper-Scissors or something (no idea how to do that yet Grin ).


Again any help is appreciated and would certainly affect the final design and final product.
Yes, this will actually be made as well. Grin

Thanks!
Sep. 04, 2013  9:21 AM
 
Do you have til June 2014 for this? I designed an improved supercharger kit for the classic minis when i did my college project so its good to see some interesting things out there.

I did engineering and got the impression that smart metal alloys weren't very cheap. If i remember correctly, aren't piezo electronic materials smart materials, could use them for sensing?
Sep. 04, 2013  9:55 AM
 
I believe that it has to be completed before the April holidays (maybe a week or two afterwards to finish anything minor).
I had quite a few ideas for the hand initially. There are a few smart materials to choose from, such as thermochromic pigments to piezoelectirc actuators/motors. piezoelectric materials can be used for sensing, however my goal is to only make the hand move for the time being, achieving at least individual finger movement, so 5 degrees of freedom.
The method used nowadays are small servo motors or things called Air Muscles (soft, lightweight linear actuators). Using servos in this project would technically simplify things, but is expensive to have small servos, as larger servos are cheap.
I thought about using an SMA, like titanol. (See this video: http://www.youtube.com/watch?v=k9f-W6Xi_Wo ) However, it probably is expensive and may not have enough strength to move a whole finger. Perhaps piezo linear actuators are a better way, but I wouldn't know.

Thanks for the help though, this is a relatively complicated project. It may have to be simplified somehow, but I have not yet gotten onto the designing phase.
Sep. 06, 2013  12:01 AM
 
I just took that survey. It was well informed and my info is really helpful.
Sep. 23, 2013  9:11 PM
 
The muscle wire looks like a good choice if you wanted something a bit more different and could be made fairly compact too, that is one funky worm too. I found this stuff to: Nitinol, bends back to its original shape with heat, I thought it was quite cool stuff, could be adapted to use like the muscle wire by bending it with a spring then using a water pipe around it to pass hot or cold water through, just a bit complicated probably for what you want to do.

How is it getting on so far?
Sep. 24, 2013  8:38 PM
 
Finishing up research stage, got a basic model hand (some tubes with string inside. Pull the string, the fingers curl up) as a proof of concept thing. The proper CAD design and modelling stage will be coming up in the next few weeks, but I do need to get some muscle wire/Nitinol in to test and see if it works with what I have in mind.
Some wires like Nitinol just need a current running through them to return to their original shape, and that way they are easily controlled. I will have to test that all out on a model finger/hand to see if it would work...

Will prob update OP with pics to show the extremely basic model and stuff.