Modeling and Control of a Tow-Lanes Isolated Crossroads

The increase in people and goods mobility during these last decades is also accompanied by many problems such as pollution, noise, traffic congestion as well as durations increasing of displacement, thus deteriorating quality of service. The crossroads is a meeting place of several lanes, complex and multipurpose. It is the most touched infrastructure by blocking and congestion situations in urban traffic. Traffic evolution is a dynamic process characterized by two kinds of behavior. Vehicles flow can be seen like a continuous system that moves along the lanes. This flow is interrupted by the traffic lights which correspond to a discrete system. It is thus there typically a hybrid dynamic system HDS. In this paper, we model an urban traffic isolated crossroads by D-elementary Hybrid Petri net. This model is translated into hybrid automaton. Each vertex of the hybrid automata is related to a D-elementary PN state. The controller has to compute the new guards (moments of lights switching) which respect imposed specifications on the continuous marking (number of vehicles on the crossroads lanes). Thus, we determine the traffic lights switching plan which optimizes flows intensity of vehicles on the crossroads lanes, and avoids the blocking and congestion situations.