Accession Number : ADA332642
Title : Intelligent Fuzzy Controller for Satellite Ground Station Applications
Descriptive Note : Final rept. Sep 95-Sep 97
Corporate Author : CLARK ATLANTA UNIV GA
Personal Author(s) : George, Roy
PDF Url : ADA332642
Report Date : SEP 1997
Pagination or Media Count : 93
Abstract : The Department of Computer Science at Clark Atlanta University investigated the use of genetic algorithms as a technique for automating the development of fuzzy logic controllers. The derivation of fuzzy controller rule-bases is relatively straightforward; however, the tuning of these controllers is a difficult process. In the first phase of this research, a genetic algorithm was used to tune the fuzzy controller. The genetic algorithm searches through the space of all membership functions to select the functions that produce the best control action. In the second phase of this project, the genetic algorithm was used to automatically derive the rule-base and membership functions. A full-featured research prototype was developed. The methodology and prototype were validated on typical control and classification problems. This research establishes a methodology for the rapid development of robust knowledge-based control systems in complex, poorly understood domains.
Descriptors : *KNOWLEDGE BASED SYSTEMS, *FUZZY LOGIC, ALGORITHMS, ADAPTIVE CONTROL SYSTEMS, RULE BASED SYSTEMS, ARTIFICIAL SATELLITES, CONTROL THEORY.
Subject Categories : Cybernetics
Distribution Statement : APPROVED FOR PUBLIC RELEASE