Harold M Hallikainen wrote:

>         Thanks for the comments. I'm also having trouble coming up with a good
>solution. This would be a replacement for Braille measuring cups. These
>exist, but the user has to stick their finger into whatever it is they
>are measuring to determine the level. It'd be much nicer if cook books
>measured everything by mass instead of volume.
>        Any more ideas?
>
Well, you could go with a stand based approach - use a stand under which
the cup is placed.  The stand can use various methods to determine the
size of the cup, and use sonar to determine the top of the material in
the cup.  Simple calculations beyond that.

Cons are that you have to 'level' the cup each time you add material if
it is dry with some shaking and place it in the stand, rather than being
told how much is in it as you pour.  It is probably more hassle than
sticking your finger it.

Alternately you'd probably have good luck if you attack it with several
types of sensors.  I'll bet that you could have a capacitive/resistive
sensor, a weight sensor, and a sonar sensor determine, together, the
type of material in the cup and how much.  It would be an expensive
approach, though.

You may be able to get away with just sonar and mass - both at the
bottom.  (or mass in the handle - could be easier to manage, since the
handle is the 1g reference point when holding the cup).

Since you are dealing with such short distances I doubt the sonar speed
is going to matter a huge amount - but you might do some tests on sonar
frequency as well - a 5 frequency analysis of the material might give
some clue as to what it is.  But even if not,  the speed of sound won't
vary a huge amount from material to material - I bet the difference is
smaller than the error of an unlevel material.

Better yet, use only a mass sensor, and a table of values for say a
dozen or two types of materials.  Put a button in the handle, and have
the user press the button -each time the unit says the current item it's
expecting.  The user simply presses the button until they get to the
right material.  Could be tedious - but a multi-button or knob interface
is exponentially more difficult to water proof.

The simple voice recognition chipsets might be of use here as well - the
user says' flour' and the cup replies "3/4 cup"

-Adam

PS  Now don't go patenting the ideas above!  They're out the the public
domain now... ;-)  Fly!  Fly free ideas!  Nest in other peoples heads!
 (grin)

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body