Generation of fuzzy numbers using maximum uncertainty possibility distributions

Abstract

A fuzzy number is used to model the information obtained from sensors. In generating the fuzzy number, the maximum uncertainty possibility distributions (MUPD) for each individual sensor measurements are linearly combined to produce the final fuzzy number representation. MUPD is the the possibility distribution which maximizes the U-uncertainty measure while constraining the optimization with the information about the structure of the distribution function. With minimal information provided, the MUPD is the most logical choice for the shape of the uncertainty distribution associated with a single sensor measurement. For sensory information, three types of information are provided: no information about the structure of the distribution; information about the first moment of the distribution; and information about the first and the second central moments of the distribution. >