Subject:
|
Re: photocell arrangement
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Thu, 20 Jun 1996 18:19:58 GMT
|
Original-From:
|
Chuck McManis <CMCMANIS@NETCOM.COMihatespam>
|
Reply-To:
|
cmcmanis@%AvoidSpam%netcom.com
|
Viewed:
|
2806 times
|
| |
|
|
About photocells...
I did this once and used some of the stuff that Rodney Brooks et al
were working on with subsumption to build the ultimate "bug". It was
a tiny cockroach type robot with a small FM wireless microphone
strapped to its back. The two behaviours it was programmed with were
"avoid light" and "find noise". Thus when someone was having a
conversation the bug would be attracted to them, but it would
stick to the shadows :-) Its definitely a conversation piece.
Unfortunately it doesn't work on carpet (it has wheels, not legs)
I found that the ideal arrangement for one direction was three
photocells. One left, one right, and one center. Using the three
samples you can plot a intensity graph and solve for the local
maximum at the center. Think of it as turning your robot until the
magnitude of the center sample exceeds the magnitude of the two
outer samples. (depending on how you have connected them this could
be a 'lower' reading or a 'higher' reading. I believe on the handy
board that "more light" is sampled as "less voltage" (ie lower
resistance)
And you should put a hood of some sort on your detectors to minimize
interference and maximize the dynamic range of the sensors. I used
two paper plates with some thin baffles in them to create a hood
that looked like: (top view)
--**--+--**--+--**--
/ L | C | R \
/ | | \
/ | | \
== | | ==
=== | | ===
================
The '=' chars are supposed to represent the curve of the edge of the
plate. (front view)
-------------------------------
|| @ | @ | @ ||
-------------------------------
The '|' are standard corrugated cardboard cut into 3/8" thick strips.
The downside of this arrangement is that the up/down visibility was
pretty low in my case but you could easily increase it by making the
hood bigger or adding vertical layers (and more photocells)
--Chuck
Jeff Keyzer wrote:
>
> I'm experimenting with photocells, and I'm looking for some feedback
> from anyone who's got some insight or experience with this sort of thing.
> Essentially, I have two photocells, both facing forward, on the left and
> right side of my robot. I'd like to use them to track light sources, such
> as a flashlight, LED, anything like that. I'm trying to figure out the best
> way to shield unwanted light away from the sensors, and right now I have
> about an inch of heatshrink tubing extending from the sensors. However, I
> have a feeling that this might be too long to yield any sensible results
> from the sensors, since the field of view of the sensors is extremely
> narrow. Does anyone know the best way to do this? My next step will
> probably be to chop some length off of the heat shrink tubing, but I'm not
> sure quite how long to make them...
--
Chuck McManis http://www.professionals.com/~cmcmanis
Director System Software, FreeGate Corp. cmcmanis@netcom.com
All opinions in the non-included text above are those of the author and
not of his employer or avocado plant.
|
|
Message is in Reply To:
 | | photocell arrangement
|
| I'm experimenting with photocells, and I'm looking for some feedback from anyone who's got some insight or experience with this sort of thing. Essentially, I have two photocells, both facing forward, on the left and right side of my robot. I'd like (...) (29 years ago, 19-Jun-96, to lugnet.robotics.handyboard)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
