Application of the Stochastic Dynamics Model for the Formation of Traffic Light Control Signals

Abstract

This article is devoted to the actual problem of traffic flow management. The purpose of the study is to optimize the congestion of the road network lanes using traffic lights with dynamically generated phases of permissive signals and percolation properties of the road network. As the main components of control, the author singles out smart traffic lights, showing the importance of the phases of traffic light regulation. The control of cars flows on each road lane used as data for management purposes. It is shown that the percolation threshold of the transport network can set a parameter that characterizes the necessary reliability of the transport network and signals the needs for additional control, if such threshold of the number of cars on the lane is exceeded. The idea of prohibiting driving into traffic lanes (after traffic lights) if the number of cars on them exceeds the allowed threshold is proposed. To perform calculations of the dynamic phases of traffic light control and to select the appropriate lane for control, the shown model of the stochastic dynamics of lane occupancy is used, in which the change in the number of cars is considered as a random process. A generated traffic control program is shown. For calculating the next permissive control signal and its duration, a road lane with suitable parameters must be selected, a traffic light control phase must be formed and a permissive signal for the selected lane and included lanes that can be opened for passage (along with the selected one). At the same time, it is considered that some of the intersection exit lanes may be inaccessible beware of the exceeded number of cars on them. The results of the comparison of the resulting dynamic traffic light and statically determined traffic light phases in the SUMO simulation system show that the described solution can reduce the number of completely blocked road lanes by 34.5%, with a slight deterioration in the overall capacity.