Re: Fuzziness: what am I missing?

Scott Dick (dick@morden.csee.usf.edu)
Mon, 6 Apr 1998 01:00:34 +0200 (MET DST)

--------------2F26EBBDDB9E67C0B870A69B
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Andy Bates wrote:

> Okay, I'm confused about Fuzzy Logic. From what I understand of it, it's
> basically an easier way of generating a function curve between inputs and
> outputs (when applied to logic circuits, for example). But the curve itself
> is still a line, not a group of patches, so why is this any more desirable
> than, say, some function that forms that line?
>
> Let's see if I can make this clear with an example. For a sample problem,
> I'll use the basic fuzzy air-conditioner problem: if hot then blast, if
> warm then normal, if cool then low. So, no matter what the fuzzy rules, for
> any given temperature (input), that input will be fuzzified, correlated
> with a fuzzy output value, then that value will be de-fuzzified into a
> specific fan speed (output). So if you drew a graph, you could match all
> input values with a corresponding specific output value, forming a curve on
> the graph. Therefore, how is this any different than, say, taking a normal
> mathematical function and mapping the input temperature to the output fan
> speed?
>
> Help me! I'm confused here. Fuzzy Logic seems great as a concept, but I'm
> having trouble figuring out how to apply it to programs, or what benefit I
> gain from it. I am trying to create a simple application using fuzzy logic
> as a sort of proof-of-concept, but I seem to be losing the faith as far as
> the application.
>
> Andy Bates.

Andy,

The trouble with finding a mathematical function is that you might

not be able to do it. Fuzzy control systems are inherently nonlinear --

and most of the success stories in implementing fuzzy control involve

nonlinear physical systems. As you doubtless know, nonlinear DEs are not

generally solvable by analytical means, and hence it is often not possible

to find an analytical function describing a nonlinear system. Fuzzy systems

provide a simple, intuitive means for creating a powerful nonlinear control

system. A good discussion of this topic can be found in Li-Xin Wang's

A Course in Fuzzy Systems and Control.

Hope this helps!

--
***********************************************************************
*                               *                                     *
*  Scott Dick                   *                                     *
*  Research Assistant           *       Cool & brilliant thought      *
*  USF Computer Science         *       still under construction       *
*  dick@morden.csee.usf.edu     *                                     *
*                               *                                     *
***********************************************************************

--------------2F26EBBDDB9E67C0B870A69B Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit

Andy Bates wrote:

Okay, I'm confused about Fuzzy Logic. From what I understand of it, it's
basically an easier way of generating a function curve between inputs and
outputs (when applied to logic circuits, for example). But the curve itself
is still a line, not a group of patches, so why is this any more desirable
than, say, some function that forms that line?

Let's see if I can make this clear with an example. For a sample problem,
I'll use the basic fuzzy air-conditioner problem: if hot then blast, if
warm then normal, if cool then low. So, no matter what the fuzzy rules, for
any given temperature (input), that input will be fuzzified, correlated
with a fuzzy output value, then that value will be de-fuzzified into a
specific fan speed (output). So if you drew a graph, you could match all
input values with a corresponding specific output value, forming a curve on
the graph. Therefore, how is this any different than, say, taking a normal
mathematical function and mapping the input temperature to the output fan
speed?

Help me! I'm confused here. Fuzzy Logic seems great as a concept, but I'm
having trouble figuring out how to apply it to programs, or what benefit I
gain from it. I am trying to create a simple application using fuzzy logic
as a sort of proof-of-concept, but I seem to be losing the faith as far as
the application.

Andy Bates.

Andy,
    The trouble with finding a mathematical function is that you might
not be able to do it. Fuzzy control systems are inherently nonlinear --
and most of the success stories in implementing fuzzy control involve
nonlinear physical systems. As you doubtless know, nonlinear DEs are not
generally solvable by analytical means, and hence it is often not possible
to find an analytical function describing a nonlinear system. Fuzzy systems
provide a simple, intuitive means for creating a powerful nonlinear control
system. A good discussion of this topic can be found in Li-Xin Wang's
A Course in Fuzzy Systems and Control.
Hope this helps!
-- 
***********************************************************************
*                               *                                     *
*  Scott Dick                   *                                     *
*  Research Assistant           *       Cool & brilliant thought      *         
*  USF Computer Science         *       still under construction       *
*  dick@morden.csee.usf.edu     *                                     *
*                               *                                     *
***********************************************************************
  --------------2F26EBBDDB9E67C0B870A69B--