Title :
Extended defuzzification methods and their properties
Author :
Runkler, Thomas A.
Author_Institution :
Corp. Res. & Dev., Siemens AG, Munich, Germany
Abstract :
Defuzzification is the selection of a significant crisp value with respect to a fuzzy set. Many applications use the standard defuzzification methods, the center of gravity or mean of maxima, to generate crisp output. These standard methods fail in some applications. In these cases, extended defuzzification methods offer an opportunity to improve the system behavior. Some extended defuzzification methods are presented, and their properties are investigated showing their advantages and preferred application fields
Keywords :
approximation theory; decision theory; fuzzy set theory; fuzzy systems; graph theory; centre of area; centroid approximation; compatibility; constraint decision defuzzification; defuzzification; fuzzy set theory; graphical analysis; histograms; membership function mapping; parametric method; section invariance; Accidents; Fuzzy control; Fuzzy sets; Fuzzy systems; Gravity; Histograms; Knowledge based systems; Message-oriented middleware; Mobile robots; Remotely operated vehicles;
Conference_Titel :
Fuzzy Systems, 1996., Proceedings of the Fifth IEEE International Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-3645-3
DOI :
10.1109/FUZZY.1996.551822