🚀 go-pugleaf

This sort of thing may've been discussed here before, but since I
don't normally read this group...

For awhile now, I've been kicking around an idea for a project.  Most
of it I either know how to do now, or can feel comfortable about hand-
waving at for now.  The only part I'm really stumped on involves a
distance sensor.  Without trying to be coy, I'll just describe what
characteristics I'd like the sensor to have, rather than describe the
whole project.

Basically, the sensor

  - must be able to measure distances of about 0 - 6", with an accuracy
    of about 1/8" or better;
  - must be non-contact (well, perhaps I should say "ideally won't
    depend on contact");
  - will be pointed at surfaces which aren't very reflective, are
    of a range of colors, are somewhat uneven (say, a variance of
    1" or less in height), aren't metallic, and cannot be marked in
    any manner beforehand (i.e.: with a tracking grid of some sort);
  - needn't worry about rapid movement (i.e.: the target will be
    stationary, and the sensor movement well-defined);
  - must be fairly cheap (say, less than $200, including the sensor
    itself and drive electronics).

I guess Polaroid sells a "hacker's kit" with the sonar gizmo they use
on their autofocus cameras, and I think for under $200.  But I seem to
recall reading that the range was more like 12" - dozens of feet, with
an accuracy of 1" or worse.  Does anyone know if that's a fundamental
limit on the sensor itself, or only the way the electronics were
designed in the kit?  (Be gentle; I'm mostly a software person.  I
know enough about electronics to think I know what I'm doing until
it's too late. ;-)

Also, I'd guess that the surface characteristics of the targets probably
rule out an infrared design?

Anyone have any ideas for me?  Am I asking for the moon with that list?

Thanks in advance!
--
>>"Aaiiyeeee!  Death from above!"<< | Steve Rehrauer, rehrauer@apollo.hp.com
   "Flee, lest we be trod upon!"    | The Apollo System Division of H.P.