home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #30 / NN_1992_30.iso / spool / sci / engr / control / 363 < prev    next >
Encoding:
Text File  |  1992-12-16  |  1.6 KB  |  31 lines
  1. Newsgroups: sci.engr.control
  2. Path: sparky!uunet!cs.utexas.edu!swrinde!emory!sol.ctr.columbia.edu!ira.uka.de!Germany.EU.net!urmel.informatik.rwth-aachen.de!solaris.rz.tu-clausthal.de!news!tw3!hueske
  3. From: hueske@iw.uni-hannover.de (Marc Hueske)
  4. Subject: Neuro-Fuzzy (Kohonen)
  5. Message-ID: <1992Dec16.121958.15425@newsserver.rrzn.uni-hannover.de>
  6. Sender: news@newsserver.rrzn.uni-hannover.de (News Service)
  7. Reply-To: hueske@iw.uni-hannover.de
  8. Organization: Institut fuer Werkstoffkunde, Uni Hannover
  9. Date: Wed, 16 Dec 1992 12:19:58 GMT
  10. Lines: 19
  11.  
  12. Help needed !
  13. I am looking for references about Kohonen's feature maps used to extract
  14. rules and membership funcitons for a fuzzy logic controller (vehicle control)
  15. from example data. Is it possible to use the feature maps to adaptively change the controller's parameters (membership functions, rules) during the operation or is the 
  16. learning ability to slow?
  17. Thanks in advance,
  18.  
  19. Marc
  20.  
  21. P.S.: I already have the report ForT2 from the University of Dortmund.
  22.  
  23. ----------------------------------------------------------------------
  24. |  Marc Hueske                    |  Tel: INT-49-511-762-3906        |
  25. |  Universitaet Hannover          |  Fax: INT-49-511-762-5245        |
  26. |  Institut fuer Werkstoffkunde   |  Email: hueske@iw.uni-hannover.de|
  27. |  Fachgruppe Unterwassertechnik  |                                  |                                       |  Appelstr. 11a                  |                                  |                                     |W-3000 Hannover, GERMANY         |                                  |                                       ----------------------------------------------------------------------
  28.  
  29.  
  30.  
  31.