Traffic Flow Dynamics Data, Models and Simulation

Author: Martin Treiber and Arne Kesting

To keep people moving in an era of increasing traffic and limited resources, science is challenged to find smart solutions. Over the past few years, research from engineers, physicists, mathematicians, and behavioral psychologists has led to a better understanding of driver behavior and vehicle traffic. This interdisciplinary field of traffic flow is set to gain even greater ground in the future. Big progress. The focus is on new applications, ranging from new driver assistance systems to intelligent methods for optimizing traffic flow, precise detection of traffic jams, and short-term prediction of traffic with dynamic navigation aids.

This textbook provides a comprehensive and didactic description of the different aspects of vehicle traffic flow dynamics and how to describe and simulate them using mathematical models. We hope to make this fascinating field accessible to a wider audience; so far it has only been documented in dedicated scientific papers and monographs.

The first part describes how to obtain and interpret traffic flow data, which is the basis of any quantitative modeling. The second and main part is devoted to methods and models used to mathematically describe traffic flows. The starting point for most models are basic concepts of physics - many particle systems, fluid mechanics and classical Newtonian mechanics - enhanced by behavioral aspects and traffic rules. On the website accompanying this book,1 readers can interactively run a selection of traffic models and reproduce some of the simulation results shown in the figures. Section 3 outlines the main applications including traffic state estimation, fuel consumption and emission modeling, determining travel times (the basis of dynamic navigation) and how to optimize traffic flow.

This book is written for engineering and transportation science students, lecturers and professionals as well as interested students. It also provides material for project work in programming, numerical methods, simulation and mathematical modeling at the college and university level. The reference implementation in the multi-model open source vehicle traffic simulator MovSim2 can be used as a starting point for readers' own simulation experiments and model development.

This work originated from the lecture notes of the Traffic Flow Dynamics and Modeling course at the Technical University of Dresden in Germany; these were previously published in the German book "Verkehrsdynamik und Simulation" by the same publisher. The English version has been updated and expanded to include new topics, for example, on model calibration. To highlight its textbook nature, it contains a number of problems with detailed solutions.

We thank all colleagues at the Department of Transport Econometrics and Modeling at the Technical University of Dresden and especially Dirk Helbing for various scientific discussions and stimulation. We would also like to thank Marietta Seifert, Christian Thiemann and Stefan Lämmer for their suggestions and corrections. Special thanks to Martin Budden, a native English speaker, for reviewing the manuscript. He is also one of the main contributors to MovSim. Finally, we would like to thank Martina Seifert, Christine and Hanskarl Treiber, Ingrid, Bernd, Dörtet Kesting, Claudia Perlitius and Ralph Germ for their valuable suggestions on this book.

Guess you like

Origin blog.csdn.net/subin0403/article/details/134667208
Recommended