Re: kosko's FAM


Subject: Re: kosko's FAM
From: Pramit Sarma (pramits@bom8.vsnl.net.in)
Date: Thu Mar 02 2000 - 16:40:52 MET


My interest is in nonlinear/intelligent control and includes the subset
of FL called FL control (FLC). This has a FAM construct of course ...
though often specialised in the 2-in/1-out (TISO) format, and hence a
2-D tabular object. The more general, Ni x Nout type of FAM is at least
very difficult to visualise even if it were not hard to compute. Higher
dimensional FRB/FAM's are preferably avoided/reduced/manipulated. So
assuming that this may still be of some interest:

TISO FAMs are special to FLC, and typical of ones in literature. In
Kosko's books there are a whole bunch of FAM's, including heuristic and
machine-generated ones. There are many implications of these things. Eg.
in a FLC, being by design a closed-loop controller, the FAM, and the
crisp equivalent input/output surface, is essentially a (fuzzy)
phase-plane, which tries to model a basic inherently stable controller,
modulo tuning and other variations. These things are quite critical to
FLC operation (Mamdani, 1976 - 1979 and on, has provided considerable
insight into this; Ray and Majumder, IEEE SMC, 1984, have formally shown
the stabilising construction of the FLC; Tang and Mulholland, IEEE SMC,
1987 show how the FLC is simply a complex PID ... as infinitum; and Lee,
IEEE SMC 1990 is a good reference paper). Again, in FLC esp., each of
these refs. will also provide considerable insight into the "classical"
TISO FLC FAM design. The meta-FAM of Procyk and Mamdani, Automatica,
1979; eg. is fascinating. For other sets of FAMs, "Industrial
Applications of Fuzzy Control", North-Holland, 1985, are particularly
useful. Otherwise, the SMC and Fuz Sys J. of IEEE are a continuous
source of information on many such things.

Pramit

Marco Mililotti wrote:
>
> Hello,
> I'm looking around for papers and resources about Fuzzy Associative
> Memory discussing the theory. Is there any source of information about the
> use of FAM in "intelligent" reasoning or something like??
>
> It would be very big help also any resource on the subject I could be
> directed at :)
>
>
> Tnx for any reply...
>
> -- Marco Mililotti
>
> ############################################################################
> 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

-- 

========================================================================== Pramit 'Jake' Sarma

E-mail: PramitS@vsnl.com [Res] PSarma@ChE.IITB.ernet.in [Lab] W-mail: pramit_s@india.com

Processes + Control, Intelligent & Simulation Systems | P R O C I S S | {Systems + Mathematics}---{Real World Processes} ==========================================================================

############################################################################ 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 - 15:59:47 MET DST