Cellular Automata Based Traffic Flow Simulation And The Optimization

Transportation plays an important role in the national economy and social development. Whether the transport system is modern, or traffic management is advanced, is an important symbol of modernization. With the development of economy, the number of vehicles increases rapidly with the demand of transportation. Since the 1980s, the growing traffic flow has produced several problems such as the traffic congestions, traffic accidents, energy shortages and environmental pollution. Thus, the study of traffic flow has become an interesting issue in the economic and social development. At present, governments and research institutions have paid significant attention to the effective use of the transport resources, as well as the application of scientific theories to guide the construction of transportation. Based on the cellular automata (CA) theory, the urban traffic is simulated, and the traffic dynamics is described by proposed discrete equations. A 2-D typical urban traffic model with four crossroads is established in this paper. Moreover, a set of factors affecting on the efficiency of traffic running, i.e., the interaction of vehicles, overtaking and signals control strategy, are investigated. The main contents and achievements in this paper are listed as follows:1) Combined the CA method with the car following model, this paper has proposed a vehicular movement model in one-way one-lane road, in which the time-position relation of vehicular movement states has been obtained. The interactions of two vehicles with different performances (acceleration ability and maximum allowable velocity), as well as the influence of the performance parameters on the movement states are studied. The traffic signals are also considered in this model, and the effect of signals control strategies (by changing the cycle time of traffic signal and the ratio of green time to cycle time) is discussed. The results show that when the signals are always green, the relative position of vehicles is only related to the performance of vehicles. When the signals change cyclically, the parameters such as signal's cycle time, the ratio of green time to cycle time, et al. could affect the movement of vehicles. Especially, in the model with considering the influences of vehicular acceleration and deceleration, increasing the waiting times of vehicles will reduce the efficiency of traffic flow.2) Based on the model proposed above, two-lane traffic flow with the consideration of overtaking is investigated. Influences of signal cycle time and vehicular density on the mean velocity and mean overtaking times of traffic flow are discussed. The effects of slow vehicles and road barricades on the traffic flow are also studied. Simulation results show that the vehicular density and the signal cycle time have significant influences on the traffic flow. The mean velocity of the traffic flow could keep a comparatively large value when vehicular density is less than the optimal density. When the vehicular density exceeds the optimal density, the state of traffic flow is affected by the density significantly. For a certain value of density, the mean velocity displays a serrated fluctuation with cycle time. Therefore, there may exist a certain combination of density and cycle time which optimizes the traffic flow efficiency. The effect of barricades on traffic flow is highly related to vehicular density. The higher the vehicular density value is, the more conspicuous the effect of barricades will be.3) The 2-D typical urban traffic with consideration of four crossroads are investigated. A parameter pcross is defined which means the allowed-density of vehicles at crossing. Influences of signal cycle time, vehicular density and allowed-density of vehicles at crossing on the mean run-time, mean velocity and mean overtaking times of the traffic flow are discussed. Simulation results show that there exist certain range ofρcross which optimizes the traffic flow efficiency, and avoids traffic jam and deadlock at crossing. The model proposed here and the simulation results which took into account the effects of signal cycle time, vehicular density, density of vehicles at crossing on the traffic flow with overtaking allowed, can reflect the situation of the traffic flow in a more realistic way.4) According to the above 2-D typical urban traffic model, the effect of two signals control strategies, i.e., the fixed-time control and the actuated control are discussed. With the fixed-time control, traffic lights turn green and red synchronously, while with the actuated control, the traffic lights change according to the number of waiting vehicles on the crossroad. Based on numerical simulation, the different effects of two signals control strategies on the traffic flow and the optimal parameters of actuated signal control are discussed. Meanwhile, the effect of the intersection spacing on traffic flow with signals control is studied. Simulation results show that the effect of signals control strategies and the intersection spacing are related to the global density of traffic flow, and there exists certain optimal parameters of actuated signal control in which the efficiency of traffic flow can be optimized.