Traffic Signal Self-organizing Control Based on Phase Random Traffic Demand

Abstract

In order to solve the problem of traffic signal control under fluctuating and unsteady traffic conditions, an urban traffic signal self-organizing control model based on phase random traffic demand is proposed. The traditional traffic signal control model based on cycle and green ratio is discretized into a real-time online control model with the phase duration as the control parameter. According to the probability density distribution model of traffic flow and the queuing state of vehicles at local intersection, the prediction model of phase random traffic demand at local intersection is established. The self-organizing control rule of phase green light duration is established with the control objective that the current phase queuing vehicles are released exactly. According to the self-organizing control rules, the phase duration of traffic signal matches the traffic demand of the current phase in real time, so as to adapt to the fluctuation of traffic flow and reduce the delay time. Simulation results show that the self-organizing control model has a significant advantage over traditional cycle-based traffic signal control under the condition of high traffic saturation.