This paper proposes a novel inference method to estimate lane-level traffic flow, time occupancy and vehicle inter-arrival time on road segments where local information could not be measured and assessed directly. The main contributions of the proposed method are 1) the ability to perform lane-level estimations of traffic flow, time occupancy and vehicle inter-arrival time and 2) the ability to adapt to different traffic regimes by assessing only microscopic traffic variables. We propose a modified Kriging estimation model which explicitly takes into account both spatial and temporal variability. Performance evaluations are conducted using real-world data under different traffic regimes and it is shown that the proposed method outperforms a Kalman filter-based approach.
Lane-level traffic estimations using microscopic traffic variables
01.09.2010
388083 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Distributed Classification of Traffic Anomalies Using Microscopic Traffic Variables
Online Contents | 2013
|