Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!cs.utexas.edu!utnut!nott!cunews!schoolnet!jlamorie
From: jlamorie@schoolnet (Joshua Lamorie)
Subject: Optic short distance range finder
Message-ID: <CynI06.M0x@cunews.carleton.ca>
Sender: news@cunews.carleton.ca (News Administrator)
Organization: Carleton University
X-Newsreader: TIN [version 1.2 PL0]
Date: Wed, 2 Nov 1994 17:31:18 GMT
Lines: 40


Gidday there,

	I don't know if this would work, or how to implement it, but 
here is the idea.  It was inspired by the Dambusters of WWII....

      _ |____________________|  <==  bottom of robot
      ^  ^^  phi           []  <--  light source at a fixed angle
      |  |  \             /
      y  |		/
      |  |    \       /
      |  |theta     /
      v  |	\ / <--- beam of light intersecting ground or object
   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  <--- ground...
         |<--x-->|    N.B..   the ^^ in the top left is a light sensor THING.

             
	OK.  Now that I have done the pretty picture.  I hope this makes 
some sense.  	What should I do with the light that hits the ground.  
Should I use the angle of the line from the receptor to the point on the 
ground where the light hits... and use some simple trig to figure out y, 
from either angle phi, or theta.

OR should I find the distance x, and use that to find y.   Which would be 
easier to implement, and do you think that this would be more accurate 
than ultrasonic?

	Perhaps a second light source where ^^ is projecting straight 
down can show two points of light on the ground, then using a CCD and 
image recognition, we could see the distance between the two points.

	Note, I am aware that this will not be that accurate at larger 
distances, but I want it specifically for short distances.  

	Oh, yeah.  Is there something that does this already?

thanks in advance

JOshua

