DPRG List  

DPRG: Contest Walls

Subject: DPRG: Contest Walls
From: Saenz, Leonel R (Leonel) lsaenz at lucent.com
Date: Tue Mar 9 10:57:07 CST 1999

The texture of my walls at home is lightly spackled (very small
knowdown).  During my home tests they were always at night with a 100
watt light bulb on with an opaque diffuser in the middle of the room.
The light is incandescent as opposed to flourescent lighting at the
contest.  I expected that beige spackled walls would reflect less IR,
scatter it more than smooth white walls.  I also suspect that the height
of the walls should not be a factor since IR is horizontally polarized.

- -----Original Message-----
>From: David Philip Anderson [mailto:dpa at io.isem.smu.edu]
Sent: Tuesday, March 09, 1999 10:06 AM
To: dprglist at dprg.org
Subject: DPRG: Contest Walls


Leonel writes:

> The walls of my test course at home were beige latex walls.  They were
> twice as reflective to IR than the DPRG course walls.  In other words,
> (my robot) would navigate the walls at home about 12 inches away with
> batteries, with the IR LED's detuned all the way.  At the DPRG course,
> would come very close to touching the walls during trial runs.
> I made the IR LED's 25% more sensitive to get about 3 inches from the
> contest walls.  My IR LED sensors sit about 4.5 inches from the floor.
> was surprised at how much the DPRG walls did not reflect IR.  You
> think that white paper covered walls would reflect mor than biege
>    covered walls.

My robot can normally see a nice white wall, cardboard box, etc at about
30 inches.  At the contest this was more like 12 inches.  However, this
was also the case for the white cloth bag that I had it dodge in the
TTime contest.  I believe the lighting in the contest room, and not the
material of the contest walls, is the culprit.  Should be easy to test.



More information about the DPRG mailing list