Simulation of a vehicle traffic control network using a fuzzy classifier system

Abstract

A Complex Adaptive System (CAS) is a network of communicating, intelligent agents where each agent adapts its behavior in order to collaborate with other agents to achieve overall system goals. Further, the overall system often exhibits emergent behavior that cannot be achieved by any proper subset of agents alone. A graphical simulation library called Operational Evaluation Modeling for Context-Sensitive Systems (OpEMCSS) has been developed to simulate complex systems, including CAS. This simulation library includes a Classifier Event Action block that is a forward chaining, expert system controller. The Classifier Event Action block can implement both crisp and fuzzy rules. A network of traffic light controller agents, one at each intersection, is simulated for a city traffic grid. Each traffic controller agent uses a fuzzy classifier block to make decisions about traffic light timing in order to minimize local vehicle wait time. Out of the co-evolutionary interaction of these agents, emerges the global minimization of vehicle wait time in the network.