Newsgroups: comp.robotics
Path: brunix!cat.cis.Brown.EDU!agate!howland.reston.ans.net!swrinde!elroy.jpl.nasa.gov!decwrl!netcomsv!netcom.com!pauljo
From: pauljo@netcom.com (Paul Johnson)
Subject: Digital D.C. Servo Math?
Message-ID: <pauljoCuqtp2.2q7@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
Date: Thu, 18 Aug 1994 18:22:14 GMT
Lines: 22

I am looking for source information (e.g. books, articles, code samples, 
etc.) on writing a digital D.C. servo (sort of <g>).  I have "invented" 
several routines of my own, all of which work, none to my satisfaction.

Normal PID math must be tuned for load, friction, etc., and has artifacts 
such as servo lag, ringing and so on.  I am looking for something to 
provide much tighter (whenever possible, perfect) control.

I have a periodic interrupt, an input position which changes on each 
interrupt, maximums for speed, acceleration and delta acceleration, and an 
output position which wants to match the input position as closely as 
possible under all conditions without violating the maximums.

I'm sure that someone much smarter that I must have attacked this problem 
already, and I would like to ride on his/her coattails.  Any one know of 
anything out there?

Thanks!
-- 
Paul Johnson                | "It aint what you don't know that'll hurt
pauljo@netcom.com           |  you, it's what you DO know that's wrong."
75470.1721@compuserve.com   | - Attribution uncertain.
