基于模糊神经网络的实时路段行程时间估计

基于对我国城市交通流的物性分析 ,提出了一种基于模糊神经网络的实时路段行程时间估计模型 ,用于将来自于交通控制中心的实时交通数据转换成为能够反映路段实时运行状况的直观参数 :路段行程时间 ,从而为交通流诱导服务 .这种方法用具有更高智能的神经网络实现了对抽象模糊规则的自动纠错的记忆 ,符合人类认识的模式 ,能令人满意地表达经验知识 ,而且模糊输入输出关系具有了明确的表达能力 .     

Modeling Car-Following Dynamics During the Starting and Stopping Process Based on a Spring System Model

Car-following models describe how one vehicle follows the preceding vehicles.ln order to better model and explain car-following dynamics,this paper categorizes the state of a traveling vehicle into three sub-processes:the starting(acceleration)process,the car-folloing process,and the stopping(deceleration)process.The stating process primarily involves vehicle acceleration behavior.The stopping process involves not only car-following behavior but also deceleration behavior.This paper regards both the stopping process and the starting process as spring systems.The car-following dynamics during the starting process and the stopping process is modeled in this paper.The parameters of the proposed models,which are represented in the form of trigonometric functions,possess explicit hysical meaning and definitive ranges.We have calibrated the model of the starting process using data form the Traffic Engineering Handbook and ob-tained reasonable results.Compared with traditional stimulus-response car-following mo