Urban Traffic Flow Prediction with Variable Cell Transmission Model

The real-time and reliable urban traffic flow prediction is the foundation of traffic management and control in urban traffic. A variable cell transmission model (VCTM) is proposed to improve the ability of the urban traffic flow prediction. Based on the analysis of the defects using the cell transmission model (CTM) to predict the urban traffic flow, two parameters are introduced into the VCTM according to the traffic flow conservation law. In the VCTM, three different types of cell connections are described by three different expressions including simple connection, merge connection, and diverge connection. Two parameters of the VCTM increase computational burden, but the cell length is adjustable which ensures that the VCTM can be widely utilized in the roads with different lengths. The simulation results indicate that the VCTM meets the requirements of the traffic flow prediction in the urban traffic.

摘要城市道路交通流预测的可靠性、实时性是实施交通管理与控制的基础。为提高城市道路交通流的预测能力，提出了可变元胞传输模型（VCTM模型）。在分析元胞传输模型（CTM模型）在城市道路交通流预测方面应用不足的基础上，根据流量守恒定律，将元胞的交通流密度和元胞长度两个参数引入CTM模型，建立了VCTM模型。VCTM模型根据路段连接、汇聚、分流等三种形式的不同特点，分别建立了元胞连接、汇聚、分流的交通流传输公式。虽然VCTM模型引入了元胞的交通流密度和元胞的长度两个参数，增加了模型求解的运算量，但克服了元胞长度必须相等的局限性，确保VCTM模型可以应用于城市道路网中的不同道路。仿真结果表明，VCTM模型能够满足城市道路交通流预测的要求。