AMC-SS Stochastic Modeling and Simulation for Traffic Flow

交通流 AMC-SS 随机建模与仿真

• 批准号: 0606807
• 负责人: Alexandros Sopasakis
• 金额: $ 9.29万
• 依托单位: University of Massachusetts Amherst
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-07-01 至 2008-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0606807&HistoricalAwards=false
• 关键词: AMC; SS; Stochastic; Modeling; Simulation

Proposal ID: 0606807PI: Alexandros SopasakisInstitution: University of Massachusetts AmherstTitle: AMC-SS Stochastic Modeling and Simulation for Traffic FlowAbstractComplex dynamical systems, such as traffic flow, are characterized by coexisting phases acting at different temporal scales. Further, random perturbations, although small, can have dramatic effects on the overall system behavior. While many physical phenomena result from the coherent but a priori unknown motion of a large number of variables most of the current models require that individual motion of the particles is resolved. Such approaches lead to difficulties in obtaining predictions that are dependable or timely. We propose a hierarchy of traffic flow models that can provide a better way of predicting traffic flow observables. Specifically we employ a novel, energy driven, stochastic noise model in order to describe the different complex phases of vehicular traffic flow. The model will be capable of reproducing the flow of traffic for the chaotic and difficult to resolve higher concentration regime.Modeling the behavior of vehicular traffic flow is a challenging, current and open problem demanding an immediate solution. The proposed traffic flow model can be a breakthrough for several cross-disciplinary traffic related research areas by enabling much more accurate and potentially faster simulation of realistic traffic streams containing a large number of interacting vehicles. The coherent description of traffic can lead to better prediction strategies with direct and indirect benefits in our energy resources, environmental and safety issues.

提案 ID：0606807PI：Alexandros Sopasakis 机构：马萨诸塞大学阿默斯特分校标题：交通流的 AMC-SS 随机建模和仿真抽象复杂的动态系统（例如交通流）的特征是在不同时间尺度上作用的共存阶段。此外，随机扰动虽然很小，但会对整个系统行为产生巨大影响。虽然许多物理现象是由大量变量的连贯但先验未知的运动引起的，但大多数当前模型需要解决粒子的单独运动。这种方法导致难以获得可靠或及时的预测。我们提出了一种交通流模型层次结构，可以提供更好的方法来预测交通流可观测值。具体来说，我们采用一种新颖的、能量驱动的随机噪声模型来描述车辆交通流的不同复杂阶段。该模型将能够重现混乱且难以解决的较高集中度情况下的交通流量。对车辆交通流行为进行建模是一个具有挑战性的当前开放问题，需要立即解决。所提出的交通流模型可以对包含大量交互车辆的现实交通流进行更准确且可能更快的模拟，从而成为多个跨学科交通相关研究领域的突破。对交通的连贯描述可以带来更好的预测策略，为我们的能源、环境和安全问题带来直接和间接的好处。