NSF/USDOT: A Non-Continuum Model of the Flow of Traffic Via Aggregation and its Application to Trip-Time Prediction

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

There is empirical evidence that information acquisition and utilization at the vehicular and highway level can be used to enhance the flow of traffic. However, to realize the potential benefits of equipping vehicles and highways with information acquiring, processing and broadcasting devices, tools are needed that can model how information at the vehicular level affects the dynamics of traffic. This project seeks to develop new and innovative theories on how to model traffic flow as a finite dimensional aggregated dynamic system. The concept of an aggregated dynamical system is as follows: at any instant in time, traffic flow parameters can be describes as a set of aggregation parameters (mean, variance, mode, median, etc.) and that these parameters can be used as to define representative vehicles in the traffic stream. This project will develop a spatially discrete model for traffic flow that will use such physical principles as conservation of vehicles and vehicle dynamics to model vehicle travel through a section of highway. The project will also examine the issue of obtaining a Eulerian (macroscopic) description for a Lagrangian (microscopic) process. The model will be used to examine how the dynamics of traffic change with the introduction of Intelligent Transportation Systems such as adaptive cruise control and dynamic message signs, both of which alter the behavior of some vehicles on the highway. The success of this research will lead to a better understanding of traffic flow and driver behavior, and will lead to more accurate predictions of the benefits of Intelligent Transportation Systems. Results of the project will be integrated into both undergraduate and graduate course.