I've just read an interesting bit about how much more engaging kids found a QRIO robot when it reacted to touches on its head, arms, and hands. Here's a video:
And here's the story:
My question is: how do its touch sensors work? Its head and arms look like hard plastic or metal. Could be that they have panels floating on microswitches, but I wonder if it's something else, like capacitance sensors or some such. Anyone have any ideas?
Thanks,
- Joe
have a look at
from
Touch SensorHead Pressure Sensitive Rubber Handle Sheet Switch Hands Sheet Switch X2 Shoulders Tact Switch X2
Jim
Thanks, that's helpful. Can anyone explain "pressure sensitive rubber"? Or, for that matter, "sheet switch"?
Thanks,
- Joe
These should be fairly self-discoverable using a Yahoo or Google search. Most work using a resistive pad and rubber that is semi-confuctive (like the kind used in remote control buttons), but others use a capacitor field.
For homebrew, a lot of people have created large area pressure sensitive plates using anti-static foam. A sheet switch is the same, except that it has a Schmitt trigger or comparator to create a definite on/off transition, rather than an analog reading.
-- Gordon
One would think, but my google-fu is failing me today. "Pressure sensitive rubber" comes up with a lot of hits, but all at the research or patent stage -- and I know from past experience that there is a lot of fairly useless nonsense among the patent records, and a lot of research articles that never lead to useful devices (though you'd have to spend a fair amount of time studying each one to figure out why).
A search for "sheet switch" produces similar results. The same term is apparently used for a lot of not-quite-the-same things, that operate on various principles.
So my question wasn't so much "how might this work" -- since you're right, Google provides many answers to that -- but rather, "how DOES this actually work, in the Qrio robot?" I find it instructive to see how real, working robots solve problems, as opposed to things in the lab or patent office that might or might not work under real-world conditions.
But apparently nobody here knows exactly what Qrio is using, which is fine. I just thought it worth a shot; I'm used to others knowing a lot more about robotics than me. :)
Yeah, I've considered that too. But that's more of a pressure sensor than a touch sensor; it wouldn't be likely to register gentle touches, but would instead require a respectable squeeze.
Best,
- Joe
Sometimes you have to get creative. So you start with terms like 'pressure sensing rubber,' then note that some of the links talk about conductive rubber, so you follow those, and eventually you get to some pretty standard technologies related to detecting "touch" by creating a flexible conductive rubber over a matrix of contact points to create individual small switches. Lo and behold, you see some of the manufacturers of these products call them membrane switches, and you can use that phrase for further searches, and so on.
From there, you see some of these work by resitance, some capacitance, and I'm sure some might work using some more esoteric approaches, like optical or even acoustic. (I can imagine a touch matrix using deformable fiber optics and a small laser to create a coherent light source. Maybe someone has already thought of this, maybe not, but that's what the Mother of Invention is all about.) If you're wanting to recreate such a "skin" yourself obviously the lower cost approaches (e.g. resistance) will be preferable.
Google may return the patents hits first but only because some of those sites have been good at spamming Google's database. Be sure to try some of the other search engines, as not all spam techniques work equally on all the engines.
-- Gordon
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.