Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: spooneythegoon on July 25, 2012, 11:50:19 AM

Title: Any Ideas How This Could Be Acheived?
Post by: spooneythegoon on July 25, 2012, 11:50:19 AM
Hi.

I am currently working on a project, and need to be able to detect the presence/absence of an object which is inside a person's closed hand, from a small distance (a minimum of about 2cm from the hand, although about 10cm would be ideal). The object to be detected must be small enough to be held hidden in a hand, and the detector needs to be small enough to be worn comfortably on the wrist (probably on a lead to power and processing held elsewhere). In fact, the smaller the better.

Does anybody know of any way I could accomplish this? I am currently looking into hall effect sensors, but don't know if the output from these could be amplified to give the required range?

Many thanks.
Title: Re: Any Ideas How This Could Be Acheived?
Post by: Mr. Ninja on July 25, 2012, 07:38:12 PM
Hi.

I am currently working on a project, and need to be able to detect the presence/absence of an object which is inside a person's hand, from a small distance (a minimum of about 2cm from the hand, although about 10cm would be ideal). The object to be detected must be small enough to be held hidden in a hand, and the detector needs to be small enough to be worn comfortably on the wrist (probably on a lead to power and processing held elsewhere). In fact, the smaller the better.

Does anybody know of any way I could accomplish this? I am currently looking into hall effect sensors, but don't know if the output from these could be amplified to give the required range?

Many thanks.

You could use two rare earth magnets (one on the wrist, one on the hand), make sure they never touch though, and encase both in a plastic housing. Then, make sure the detector magnet enclosure includes a plastic rod on both ends to tactile switches. And there you have it.
Title: Re: Any Ideas How This Could Be Acheived?
Post by: Soeren on July 26, 2012, 10:14:43 AM
Hi,

[...] an object which is inside a person's closed hand,
Far too little info!
If the object is is a ball of cotton wool, a marble, a bit of plastic or similar, it's either impossible or can only be done with Roentgen.
So, please describe all in minute detail, although it may seem so, we're not oracles ;)


Does anybody know of any way I could accomplish this? I am currently looking into hall effect sensors, but don't know if the output from these could be amplified to give the required range?
That won't work with cotton wool for sure, but for magnetics, it's easy to make a Hall switch very sensitive, although it may not be the best option - more after the infusion of crucial info ;D
Title: Re: Any Ideas How This Could Be Acheived?
Post by: spooneythegoon on July 26, 2012, 11:02:05 AM
Sorry about that, I should have made it clearer. In my spare time I also dabble in magic tricks, and this is being built as part of a trick I am working on. Basically, the "object" will be a ball of clay or poker chip or something, with whatever I am detecting hidden inside. The detected item can be a magnet, RFID tag or whatever else is  required as long as it will fit.

P.S: I am aware that this may not be considered a robotics project as it doesn't move, but at least it will be connected to a microcontroller and have some level of "intelligence" in that it will make decisions and control an output, so it is kind  of a robotics project ;).
Title: Re: Any Ideas How This Could Be Acheived?
Post by: Soeren on July 26, 2012, 05:49:44 PM
Hi,

Sorry about that, I should have made it clearer. In my spare time I also dabble in magic tricks, and this is being built as part of a trick I am working on. Basically, the "object" will be a ball of clay or poker chip or something, with whatever I am detecting hidden inside. The detected item can be a magnet, RFID tag or whatever else is  required as long as it will fit.
Might be hard hiding anything inside a poker chip, unless you have a router table to take two chips to half thickness and gouge out some room for embedding, then gluing the halves together invisibly afterwards.

A magnet would spoil the fun, as it would need to be fairly large (considering you wanna hide it) and the magnet may be easy to detect if it comes near a key ring or similar - I wear a ring made out of wolfram (tungsten), so I detect any magnets I get in my hand  :P

I'd choose RFID of the highest frequency used (to decrease the size of the antenna coil) and using an "intelligent" RFID a.k.a. a transponder, each object could have its own signature - tell the victims to take one object out of several, put the rest in a box without you looking and then you can amaze them, not only by telling who's got it, but also by "deducting" which object they have.
The controller could be strapped to your body and signal by a number of small electric pulses (like a miniature TENS unit, either just a number of "beeps", or one up, a sort of Morse code like signalling. Just 3 bits of long and short periods (for "1" and "0" gives you 8 different objects identified.

Perhaps having the coil in your sleave will be enough, but you could have it inside your hand covered with artificial skin (a sort of bandage that resembles skin - chat up a nurse on your local hospital for a sample.


P.S: I am aware that this may not be considered a robotics project as it doesn't move, but at least it will be connected to a microcontroller and have some level of "intelligence" in that it will make decisions and control an output, so it is kind  of a robotics project ;).
No, however you flip it, it has got nothing to do with robotics at all.
A simple calculator or even one of the cheapest bicycle LED lamps have a microcontroller, makes decisions and have one or more outputs (as well as input(s)), but you wouldn't call them robots either (I hope).

A non robotics project of enough interest may get help (hey, I find it much less offensive than those who repeatedly post software questions in the electronics forum with the words "sorry if this is misplaced" rather than just thinking for 5 seconds before hitting the "send" button, but please don't try to con it through under the almost-a-robot label - we're not that stupid... And if we were, you wouldn't ask us ;)
Title: Re: Any Ideas How This Could Be Acheived?
Post by: spooneythegoon on August 06, 2012, 08:34:24 AM
Thanks for the help! Sorry about the "robot" thing, I wasn't trying to trick anyone.

Would anybody be able to tell me whether this module http://proto-pic.co.uk/125khz-rfid-module-uart/ (http://proto-pic.co.uk/125khz-rfid-module-uart/) would be suitable for use with a picaxe microcontroller?

Many thanks.
Title: Re: Any Ideas How This Could Be Acheived?
Post by: waltr on August 06, 2012, 06:30:47 PM
If the PICAXE has a TTL asynchronous serial interface (UART) then it is compatible.
Title: Re: Any Ideas How This Could Be Acheived?
Post by: spooneythegoon on August 08, 2012, 05:28:56 AM
Thanks! :D