CAREER: Improved Continuum Models of Vehicular Traffic Flow

PI：Michael H.Zhang，加州大学戴维斯分校，提案编号：9984239提案标题：改进的车辆交通流连续模型项目摘要：本研究旨在扩大对车辆交通流中各种关键现象的理解，并发展改进的理论来对其进行建模。交通流由不同智能粒子组成，由不同的人驾驶，呈现出复杂的动态模式。虽然交通流的某些方面，如冲击波，已经被很好地理解并支持众所周知的运动波交通流理论，但对交通流应用至关重要的许多其他现象还没有被很好地理解。这些包括层流到走走停停的流动过渡，停止-开始波，以及幻象交通拥堵。本研究建议1)更好地理解上述关键现象的本质，2)发展改进的交通理论来模拟这些现象，3)应用所开发的理论来更好地管理交通拥堵和减少车辆排放。在这项研究中，仔细审查了实验证据，仔细审查了交通流理论的过去发展，以获得对驾驶员行为的洞察和新理论发展的可能方向。然后，基于对驾驶员行为和现有理论的合理理解，改进的理论被开发出来，并通过实验数据进行了严格的分析和仔细的验证。这项研究的成功可以使人们更好地了解交通流，并极大地提高对交通流进行监测、建模和控制的能力。这些进展反过来又导致更有效地利用现有的道路基础设施和更好的空气质量规划。

PI: Michael H. Zhang, University of California, Davis Proposal Number: 9984239 Proposal Title:Improved Continuum Models of Vehicular Traffic Flow Project Abstract: This research seeks to expand the understanding of various critical phenomena in vehicular traffic flow and develops improved theories to model them. Comprising of heterogeneous intelligent "particles" -vehicles driven by diverse people, traffic flow exhibits complex dynamic patterns. Although some aspect of traffic flow, such as shock waves, is well understood and supports the well known traffic flow theory of kinematic waves, many other phenomena that are critical to traffic flow applications are not well understood. These include the laminar flow to stop-and-go flow transitions, stop-start waves, and phantom traffic jams. This research proposes to 1) better understand the nature of the aforementioned critical phenomena, 2) develop improved traffic theories to model these phenomena, and 3) apply the developed theories to better manage traffic congestion and reduce vehicle emissions. In the research, experimental evidence is carefully examined and past developments of traffic flow theory are scrutinized to gain insights on driver behavior and possible directions of new theory development. Improved theories are then developed based on a sound understanding of both driver behavior and existing theories, rigorously analyzed and carefully validated with experimental data. The success of this research can lead to a better understanding of traffic flow and greatly increase the ability to monitor, model and control traffic flow. These advances in turn lead to more efficient use of existing road infrastructure and better air quality planning.