Research of the car-following model is always of great interest. Recent studies have indicated that drivers’ following behavior varies when the lead vehicle is a passenger car as opposed to a heavy vehicle. In the age of big data, this paper presents a data-driven car-following model that is able to capture the characteristics of driving behavior contained in the field data. By using the Support Vector Regression approach and incorporating vehicle-type dependent parameters, the data-driven model is capable of reproducing car-following behaviors of multiple vehicle types with high prediction accuracy. Three vehicle-type combinations are considered in this paper: car-following-car, car-following-truck, truck-following-car. The data analysis results confirm that there exists a strong correlation between driving behavior and vehicle type and the significant differences are revealed in desired space headway and desired speed among different vehicle-type combinations. These findings provide insight into driving behavior in mixed traffic flow and illustrate the necessity of taking vehicle types into account in microscopic traffic simulation.
A Data-Driven Car-Following Model that Considers Impacts of Car-Truck Combinations
Smart Innovation, Systems and Technologies
Advances in Smart Vehicular Technology, Transportation, Communication and Applications ; Kapitel : 24 ; 265-285
2021-07-02
21 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Qld considers truck safety inquiry
Online Contents | 2008
SOUNDOFF - Brian Spratt considers the best way to source truck parts
Online Contents | 2000
Horsepower Considerations for trucks and truck combinations
TIBKAT | 1978
|Transportation Research Record | 2018
|