Re: Justification for FLC


Subject: Re: Justification for FLC
From: Edgar Dohmann (edohmann@ortech-engr.com)
Date: Thu Jan 13 2000 - 17:36:42 MET


In article <85jilt$5ro$1@nnrp1.deja.com>, pandeyashish@my-deja.com wrote:

> I would like to explain what I said in my previous post about
> justification of FLC.
>
> Why should someone go for FLC if for example a mathematical model
> exists and controller can be designed using BODE in frequency domain?
>

And my point is that there could be many reasonable justifications for
such a decision. For example, if you have an excellent GUI tool like
TILShell available for doing your FLC design, development, and testing, it
may be much faster and easier to implement the solution with FLC than for
some other, possibly equally suitable, solution.

> From what I understand from so maney replies is that FLC may not be
> used in an application where instability could lead to loss of life and
> limb but because of its marketability FLC may be applied in all other
> application not because their stability is proven but because they just
> sound cute.
>

I doubt if many serious application engineers utilize any solution "just
because it sounds cute". I know of many practical industrial and
commercial applications that have been implemented by a number of
different engineers in various companies utilizing FLC. None of these
chose FLC for their solution because it sounded "cute".

It sounds to me like you have made up your mind that FLC is worthless and
are just attacking anyone who wants to use it as making a silly decision.
If that's the case, then you really don't have anything worthwhile to
contribute to this list.

> 90% potential improvement is claimed by some for FLC(I hope you have
> guessed what it means)Well my quastion is how difficult is it to achive
> that? I am sure that its a tedious hit and trial.
>

Not necessarily if you have good development tools like TILShell.

> All I have been saying is that linear-->non-linear-->adaptive and then
> FLC may be considered.
>

Well, a good engineer should always consider various possible solutions
for a problem and select the one that best suits the situation at hand.
There are almost always many factors that lead one to the particular best
solution for any given problem and constraints. So, while I would agree
that all 4 approaches you list are valid considerations for a given
control logic design, it is rather shortsighted and narrow-minded to say
that they must be considered in the order that you listed them with the
implication that FLC is only suitable if the other 3 solutions will not
work.

>
> I hope you have heard about a plane crash in 70s which sealed the fate
> of Adaptive Control. FLCs somehow might meet the same fate.
>
> In academics there is lot of debate going on about justification of FLC
> and all I am trying to do is to get involved.
>

My point exactly. A purely academic argument that does not consider the
realities of a practical commercial project and design implementation
doesn't carry much weight in my opinion.

Edgar Dohmann

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



This archive was generated by hypermail 2b25 : Thu Apr 06 2000 - 15:59:38 MET DST