交通诱导系统智能化方案及其仿真研究

交通流过程是不确定性很强的复杂过程，用传统的理论方法已经不可能有效解决交通流控制与诱导问题，交通诱导系统（route guidance system,RGS)必须智能化，该文评述了基于动态交通分配模型的交通诱导方法，指出其存在着模型理想化、局限于解析模型和忽视了人的参与与作用等问题，提出了RGS的一种智能化方案，介绍了基于学习机制来实现交通流短时预测和推理得出交通诱导信息的原理，给出了交通诱导智能系统的结构，用仿真研究的结果说明了原理的有效性，得出了交通诱导智能化的方案比依赖某种确定的数学模型的方案更可行有效的结论。

Design of intelligent route guidance system and it's simulation

Route guidance system (RGS) must be intelligent because traffic flow process is a complex process with the serious uncertainty. It is impossible that the problems of traffic control and route guidance be solved using the traditional approaches. In this paper, the RGS approaches based on dynamic traffic assignment are reviewed. The shortcoming of these approaches lies in the idealized model, confinement of mathematical model, and ignored the human factor. A design of intelligent RGS is presented. The principle and the structure of the intelligent RGS are introduced. The short term forecasting of traffic flow and the route guidance information of the system are obtained using the principle of machine learning and inferencing engine. The effectiveness of the systems is shown by the simulation results. Finally, the conclusion is given: the intelligent RGS is more effective and feasible than the RGS based on mathematical model.