Actions for Expert system training and control based on the fuzzy relation matrix

Email RIS file
Bookmark
Report an Issue

Expert system training and control based on the fuzzy relation matrix

Author
Sheridan, T. B.
Published
Jun 1, 1991.
Physical Description
1 electronic document
Additional Creators
Ren, Jie
Online Version

Availability

I Want It
I Want It

Finding items...

Restrictions on Access
Unclassified, Unlimited, Publicly available.
Free-to-read Unrestricted online access
Summary
Fuzzy knowledge, that for which the terms of reference are not crisp but overlapped, seems to characterize human expertise. This can be shown from the fact that an experienced human operator can control some complex plants better than a computer can. Proposed here is fuzzy theory to build a fuzzy expert relation matrix (FERM) from given rules or/and examples, either in linguistic terms or in numerical values to mimic human processes of perception and decision making. The knowledge base is codified in terms of many implicit fuzzy rules. Fuzzy knowledge thus codified may also be compared with explicit rules specified by a human expert. It can also provide a basis for modeling the human operator and allow comparison of what a human operator says to what he does in practice. Two experiments were performed. In the first, control of liquid in a tank, demonstrates how the FERM knowledge base is elicited and trained. The other shows how to use a FERM, build up from linguistic rules, and to control an inverted pendulum without a dynamic model.
Other Subject(s)
Collection
NASA Technical Reports Server (NTRS) Collection.
Note
Document ID: 19910016327.
Accession ID: 91N25641.
NASA-CR-188605.
NAS 1.26:188605.
Terms of Use and Reproduction
No Copyright.
View MARC record | catkey: 15680463