Traffic Simulations with Empirical Data: How to Replace Missing Traffic Flows?
| Autor: | Lars Habel, Alejandro Molina, Kristian Kersting, Thomas Zaksek, Michael Schreckenberg |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Empirical data
Traffic congestion reconstruction with Kerner's three-phase theory Computer science Exponential smoothing Physik (inkl. Astronomie) computer.software_genre Poisson distribution symbols.namesake symbols Data mining Traffic generation model computer Transmission errors Slip (vehicle dynamics) |
| Zdroj: | Traffic and Granular Flow '15 ISBN: 9783319334813 |
| DOI: | 10.1007/978-3-319-33482-0_62 |
| Popis: | For the real-time microscopic simulation of traffic on a real-world road network, a continuous input stream of empirical data from different locations is usually needed to achieve good results. Traffic flows for example are needed to properly simulate the influence of slip roads and motorway exits. However, quality and reliability of empirical traffic data is sometimes a problem for example because of damaged detectors, transmission errors or simply lane diversions at road works. In this contribution, we attempt to close those data gaps of missing traffic flows with processed historical traffic data. Therefore, we compare a temporal approach based on exponential smoothing with a data-driven approach based on Poisson Dependency Networks. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|