|
[DPRG] Borenstein's UMBMark
Subject: [DPRG] Borenstein's UMBMark
From: Randy M. Dumse
rmd at newmicros.com
Date: Tue May 20 16:11:47 CDT 2008
> but this will perhaps become more obvious when you run the
> exercise yourself. It's a little hard to explain (obviously
;)
Back when Bill James and I were working on BeerBot, think that
was 2003 or so, we were doing Borenstein's squares within a few
inches.
I was always a bit upset we didn't do better square corners. I
could drive over a point, but I never worked out how to arrive
at a point by coming from a direction, as opposed to driving in,
and having a little turn out to go to the next point.
We did work out a way to do a controlled turn if the point was
"behind the beams" of differential turning before starting out,
but there was still that bobble as it came away with a little
bit of lateral displacement right after the turn.
> Or you might try different tires. jBot's are the "Road Rage"
> after market tires:
Yes, maybe those tires would do the trick. I couldn't remember
what you'd used on jBot. Clearly those have a better defined
"center ridge" to ensure lightly loaded contact is in the center
of the tire.
I was intrigued by the solution one of the younger members used
at the indoor contest. He put rubber bands around his knobby
tires, so the actual contact with the floor was limited to a
small patch no wider than the rubber band when on a smooth
floor.
Of course the interesting question then, which you and I have
discussed before, is how far out do you put the rubber band, to
be closest possible to the natural wheel-base width you'd get
with the native tires if you're running multiple wheels like
jBot, being closer to the outside diagonal, than the center to
center measurement.
Randy
More information about the DPRG mailing list
|
