Statistische Analyse von Verkehrsdaten und die Modellierung von Verkehrsfluss mittels zellularer Automaten

Dateibereich 5028

11,70 MB in 12 Dateien, zuletzt geändert am 12.04.2007

Dateiliste / Details

DateiDateien geändert amGröße
anhang.pdf23.05.2000 00:00:00160,3 KB
index.html12.04.2007 10:34:498,8 KB
inhalt.htm23.05.2000 00:00:001,6 KB
inhalt.pdf23.05.2000 00:00:0052,2 KB
kap1.pdf23.05.2000 00:00:0055,3 KB
kap2.pdf23.05.2000 00:00:003,27 MB
kap3.pdf23.05.2000 00:00:00269,7 KB
kap4.pdf23.05.2000 00:00:002,14 MB
kap5.pdf23.05.2000 00:00:0056,7 KB
literatur.pdf23.05.2000 00:00:00103 KB
neubert.pdf23.05.2000 00:00:005,58 MB
titel.pdf23.05.2000 00:00:0030,7 KB
Past and recent investigations of traffic dynamics mostly rest on averaged data taken over a short period of time, e.g. one minute. They make a qualitative distinction between the generally accepted traffic states called "free flow", "synchronised flow" and "Stop-and-Go traffic" possible. However, the essential car-car-interactions on the microscopic level are concealed. With extensive statistical examinations of single-vehicle traffic data presented in this work one gains new insights especially from a microscopic point of view. By means of time series analyses, correlation functions and by use of histogram methods new evaluation methods for driving behaviour are introduced. These quantities show a strong dependency on the present traffic state, the observed interval of density and the environment (e.g., freeway or city traffic). Microscopic features like synchronisation of velocities ranging over a number of vehicles or decreasing time headways smaller than one second noticeably influence macroscopic proper ties of traffic as expressed in fundamental diagrams. Moreover, on the basis of cross correlation functions connecting flow and density some quantities can be defined to discriminate between the several traffic states quantitatively. These empirical results have impacts on modelling and simulation of traffic flow. A modified cellular automaton comprising aspects of anticipation is introduced and discussed in great detail. Metastable states can be formed and all other criteria are also met to indicate a well-operating traffic flow model. A lot of simulations were done in order to analyse single-lane and multi-lane behaviour and to find out the specific problems coming along with event-driven measuremens methods using "inductive loops". Another focus are density waves to estimate their speed and to elucidate the phase separation and the transition between different traffic states.
PURL / DOI:
Lesezeichen:
Permalink | Teilen/Speichern
Dokumententyp:
Wissenschaftliche Abschlussarbeiten » Dissertation
Fakultät / Institut:
Fakultät für Physik
Dewey Dezimal-Klassifikation:
500 Naturwissenschaften und Mathematik » 530 Physik
Stichwörter:
synchronised traffic, cellular automation, fundamental diagram, traffic flow simulation, Road traffic, traffic state
Beitragender:
Prof. Dr. rer. nat. Schreckenberg, Michael [Gutachter(in), Rezensent(in)]
Sprache:
Deutsch
Kollektion / Status:
Dissertationen / Dokument veröffentlicht
Datum der Promotion:
23.05.2000
Dokument erstellt am:
23.05.2000
Dateien geändert am:
12.04.2007
Medientyp:
Text