I want to make flexible flooring system which make up of small tiles which are connected together.
What I am interested in is which tile is the people standing on, the amount of pressure apply no of interest here.
I thought if I use one pressure sensor for each tile could do the trick but it would result in a large number of input which is not preferable.
So I thought I should use something like a grid where I sense as which row and which column is being activated.
The problem is that I have no idea how to do this, so please help !!!!
Another thing is that is it possible to do something like click-on where you don't need to specify how many tiles before hand but simply connect it and make the system automatically realise how many tiles are connected in each specific row/column.
How many tiles are you going to use and how large is each tile? If it's too many, you might be better off with an overhead infrared camera and blob tracking.
Actual dimension would be something like 50x50 cm but I think I will make a 2.5x2.5 cm tile just for simulation and proof of concept.
I think 10x5 tiles would be nice.
I would like to stick to tile/floor based as much as possible though.
Ahha, I will have a look at those example.
But it is more on the coding, no?
The problem I have is more on the building of the hardware though.
And do you have any idea on how make it as "click-and-ready to use" like user don't need to reprogram but work even when the tile increase/decrease in number.
I think that wuld be very difficult, at least with the matrix way of doing it, since the software needs to know how many rows / coloumns you have to do the scanning.
Really
I am actually thinking about putting resistance for example 10 ohm in each tile and in order to know how many tiles is in the row I would use something like total resistance or something like that.
Do you think this will work ?
You could embed photoresistors in each tile and then test the resistance of the photoresistors th see which is covered by a foot and which are not (assuming the tiles are in a lighted area).
Well the main function would be to know exactly which one is being activated or stood on.
but I also would like to have a functionality where user can add more tiles in without the need to reconfigure the software, this is where I think the resistor thing can come in handy. But this is more like a secondary thing.
I would prefer if the sensitivity is all the way to the edge, however, the prototype I am going to make is more like a proof of concept so I don't really need to worry too much about it. I need to be concern more on the interaction design and stuff like that ;).
My concept is related to machine safety within the manufacturing factory/plant. I want to make a flexible area monitoring system so I am expecting the weight of an adult. But I am going to make a miniature scale simulation so I think I would be using finger to simulate person working.
Actually I also want to put lighting system within the tile so that it could flag the status of where it is dangerous and stuff. I thought about using projector but it is not so suitable in my case. But this would come in later I guess. First I need to figure out how I could do the detection part.
Okie, I will try and see how it work out then.
Still sure not whether I should connect them so rows/columns to reduce the number of inputs or not though.
I understand that before such system can be used it also need to be approved and certify but what I am doing now is more of a conceptual proposal (university term project kind of thing) for a safety equipment manufacturing company. They said not to worry about that but just think of someone interesting for them.
The scheme with the resistors would tell you only how MANY tiles in a particular row were activated, not WHICH ones (as your original problem statement asked for.)
The resistors would work if each tile is polled individually, not on an x/y matrix. Switches are probably a poor solution as they usually require some type of movement to actuate and a suspension system to accomidate the required switch movement. A patent search might supply many ideas. Physical barriers and restraints use to be popular. People are quite clever in defeating safety interlocks and guards on equipment, quite often cutting off their fingers and such as the result. Probably more info needed on the particular gizmo being operated.
An arbitrarily extend-able matrix can be managed by using shift-registers to selectively energize one axis, and shift registers along the orthogonal edge to receive the scanned signals. Diodes at each intersection would allow ALL the tiles to be sensed individually regardless of how many were activated. This is a conventional keyboard/matrix scanning scheme. Using shift registers would make it modular and extendable to any practical size.
This paragraph sound really promising but sound complicate to a electronic/electrical newbie like me. I will take sometime to digest and also maybe ask some lab technician at the university for the implementation detail ;D.
Well, you need something. Have you "assumed" some type of detection component into the tile, such as a pressure switch?
Well actually what I "assumed" would be either pressure switch or RFID reader with multiple antenna align in grid-like manner. Something like what they did here http://www.serioustoys.com/en/more_tagtiles.aspx
The rf tag idea is somewhat passive, which is a good idea, but would probably require tagged shoes or shoe covers. This would be a "two part" system which would have a lot of overhead in real world use. If you could find normally closed version of switches similar in size to the tactile switches like below, you could fairly easily make a proof of concept setup. Make tiles from plywood and put the small switches on the bottom of the tiles. Put the switches on a tile in series such that if only one switch is depressed the circuit on the tile is broken. Use multiplexing to check the current flowpath thru each tile to see if it has beeb stepped on. Disclaimer: Tactile switches are not made to be stepped on. If you don't want the expense and hassle of the contact switches, you could do ~the same using aluminum foil, thumtacks,and some soft foam.
Another request, if I want to add some kind of lighting system on top of each tile and I want to control them individually. Is there a way to minimize the wiring (not one tile to one output socket)?
I haven't start building up the prototype yet though, will be making in next (next?) week since now I am pretty caught up with other work. But it would be great to have things work out before starting to build it.
but sound complicate to a electronic/electrical newbie like me. I will take sometime to digest and also maybe ask some lab technician at the university for the implementation detail ;D.
and help to come