A. Flemming, R. Kempf, and J. Adamy (Germany)
Recurrent fuzzy systems, trafﬁc simulation, automata.
In this paper, we propose a model for microscopic trafﬁc
simulation, which uses recurrent fuzzy systems. Its rule
base reﬂects the recurrent decisions of the vehicle drivers.
In contrast to cellular automata models, which also work
rule based, our proposed model processes vague or ”fuzzy”
data and utilizes full intervals for the state values, e.g. for
the velocities. We show that the recurrent fuzzy system
is capable of modeling trafﬁc ﬂow that exhibits realistic
macroscopic behavior and where the vehicle velocities are
no longer conﬁned to a small set of discrete values on the