Subject:
|
Homebrew rotation sensor idea
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Sun, 21 Nov 1999 17:40:56 GMT
|
Original-From:
|
Dave Johnson <djohnson@%AntiSpam%sirius.com>
|
Viewed:
|
554 times
|
| |
|
|
Hi All,
New guy here (some background below, if you're interested), been lurking
for a week or so, and I'm off to buy my 1st Mindstorms set as soon as
Toys 'R Us opens this morning...they have both 1.0 and 1.5 on the shelf,
I'll be getting 1.0 after reading some past posts about the upgrade deal.
I've been reading the web avidly for several weeks now, and I'm well
fired up :-)
But to the point: Last night I went and saw "Dial M for Murder" in 3D,
and now have several pairs of polarized glasses...
As most of you probably know, if you layer 2 polarized films on top of
each other, the light transmission through them varies (linearly?) from
"clear" to opaque as one is turned through 90 degrees with respect to the
other. I have thoughts (still a bit vague) of making a rotation sensor
with polarized film and a light sensor (and perhaps a uniform light
source).
Awkward, perhaps, but depending on the sensitivity of the light sensor
you use, it seems to me it might provide pretty good resolution, probably
much better than the 22.5 degress of lego's rotation sensor. You would
have to track history to know where you were in an absolute sense (the
pattern inverts every 90 degrees, repeats every 180).
I suspect you could package the whole thing neatly in a few bricks, if
you used an led light source and made it a powered sensor.
Any thoughts on this? Anyone tried anything similar? Any other uses for
polarized film you can think of?
Dave Johnson
-------
Background:
Robotics: Been dabbling for years, and have even won 2 robotics
competitions. (But don't read too much into that: for the 1987 San
Francisco Robot Olympics I built a simple line follower but mine was the
ONLY autonomous entry, and the scoring was slanted heavily in favor of
autonomy. The second was a lego robotics competition at Alife II put on
by Fred Martin, and we won - I was working with a partner, Eric Cooper -
simply because our 'bot completed all three tasks successfully. But the
robot itself, frankly, was boring, so utilitarian as to be yawningly
dull. If there would have been scoring for creativity, we'd have been
near the bottom, there were some truly amazing creations there.)
Electronics: Lots of classes in college, lots of dabbling myself, and for
the last year have been a real live hardware guy, designing handheld data
acquisition systems for Palm-based computers.
Mechanics: Deep love of all things mechanical all my life. I was a
mechanical guy for the ILM creature shop in a former life, but I own no
machine tools of my own (something I hope to rectify someday: those
Sherline mills look lovely.)
Computing: Been a programmer since the early 80's in college (FORTH,
Fortran), but personal computers didn't do anything for me at all until I
saw a Macintosh for the first time. I've been an avid Mac programmer
since 1985 (mostly C and C++ the last 8 years or so), and worked at Apple
for many many years in Developer Support. Also quite a bit of assembly
language experience, especially on 80xx in college, and on PICs over the
last year, for work.
Lego: Total Newbie, but I'm hoping that will change quickly.
|
|
Message has 2 Replies:
 | | Re: Homebrew rotation sensor idea
|
| Excellent idea! I'd suspect that the relationship would be sinusoidal rather than linear, but who knows? If it is, it's a sine/cosine generator, too. Sone LEDs (laser diodes, don't know about others) are polarized in some specific plane. It is (...) (25 years ago, 21-Nov-99, to lugnet.robotics)
| | | Re: Homebrew rotation sensor idea
|
| Welcome aboard, Dave (and David). Sounds like a pretty good idea, but one caveats: if you plan on using the standard light sensor, be aware that it does not (in my experience) give constant readings: i.e., point it at a spot on a wall and let it (...) (25 years ago, 21-Nov-99, to lugnet.robotics)
|
5 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
