Consensus of agents with fuzzy state

The problem of distributed consensus of several clashing pieces of information has been studied for many purposes, such as for distributed estimation or mobile robot coordination problems. In this paper we apply this framework to a criticality estimation problem. Specifically, several human observers, each with a partial, incomplete and subjective view, are in charge to evaluate the severity of a failure in an highly interdependent critical infrastructure scenario. In order to cope with the subjectivity and the ambiguity of human reasoning, in this paper the classical distributed consensus problem is extended in order to handle ambiguous opinions modeled as fuzzy numbers. To this end, in this paper some convergence conditions are given, showing that the stability does not depend on the fuzzyness of the opinions.