Modelling and Analysis of Oversaturated Intersections Using Jointly Hybrid Petri net and Hybrid Automata

Abstract

The Urban traffic network is a typical example of a complex system called hybrid dynamic system. It is also a multivariable and multi-scale system having the interaction of two dynamics kinds. The difficulties encountered in the study of traffic and especially in the traffic signal control problem is how to model the traffic flow, how to define its variables and how to analyze their behavior. In this paper, we propose a modeling of oversaturated intersections traffic a view to designing a traffic control approach. This concept is based on the joint use of the two well-known hybrid representation tools: hybrid Petri networks and hybrid automata. The interest of this combination is to profit from the both models advantages while avoiding their disadvantages. A formal verification property is performed to refine this model. This technique is based on the computation of the reachable state spaces. Indeed, the new model captures important aspects of the traffic flow dynamics. Its favorable structure can be used in order to provide an efficient signal-timing plan to avoid oversaturation and to ease congestion. The numerical results show that the coordination traffic signal obtained by the proposed control approach outperforms those obtained using the widely utilized signal timing optimization software SYNCHRO under various demand scenarios from unsaturated to oversaturated.