|
[DPRG] Absolute reference for angular position
Subject: [DPRG] Absolute reference for angular position
From: Sluggy!
sluggy9912 at swbell.net
Date: Tue May 27 14:16:58 CDT 2008
All the solutions given up to this point seem to be
towards the robot being self-calibrating. The only
problem with any of these is that it's adding another
sensor, one that is intentionally narrow in focus,
with all the problems that represents.
If this measurement is for one time (or even
occasional) manual calibration, how about a laser
pointer attached to the chassis and pointed to a spot
on the wall. Mark that spot. Rotate the robot under
program control to what it thinks is 360 degrees.
Using a high tech ruler, measure the error by where
the laser ended up pointing. Repeat many times,
perhaps averaging the results. Adjust the count in the
software.
Sluggy!
---
"The idea that unicycling is intrinsically funny does not explain the findings." Professor Sam Shuster.
More information about the DPRG mailing list
|
