Newsgroups: comp.robotics
Path: brunix!news.Brown.EDU!noc.near.net!howland.reston.ans.net!usc!sdd.hp.com!hpscit.sc.hp.com!hplextra!hplabs!hp-cv!hp-pcd!hpcvca!hpcvcec.cv.hp.com!abrown
From: abrown@hpcvcec.cv.hp.com (Allen Brown)
Subject: Re: Fuzzy Logic ?
Message-ID: <1993Apr21.012325.10224@hpcvca.cv.hp.com>
Sender: nobody@hpcvca.cv.hp.com
Nntp-Posting-Host: hpcvcec.cv.hp.com
Organization: Hewlett-Packard Company, Corvallis, Oregon USA
References: <nagleC5t1y7.GAv@netcom.com>
Date: Wed, 21 Apr 1993 01:23:25 GMT
Lines: 20

: abrown@hpcvcec.cv.hp.com (Allen Brown) writes:
: >Another thing that you don't mention (and the last guy completely
: >missed) is that the solutions come out first order continuous even tho
: >you may have entered a discontinuous description of the problem
: >domain.
:        The derivatives have discontinuities, though.
: 						John Nagle

With triangular membership functions I can see this is true.  But if
the membership function is continuous in its derivatives the fuzzy
solutions should also be.  Of course this is more costly in run time.
But it is something the programmer doesn't need to worry about except
to choose the right membership function.  On the other hand, the fuzzy
library programmer does need to worry about how to make a continuous
membership function.

[Said by someone who's knowledge of fuzzy comes from brochures.]
--
  Allen C. Brown  abrown@cv.hp.com or hplabs!hpcvca!abrown or "Hey you!"
      Not representing my employer.
