Including heavy vehicles in a car‐following model: modelling, calibrating and validating

Abstract

Summary Heavy vehicles influence general traffic in many different ways compared with passenger vehicles, and this may result in different levels of traffic instability. Increases in the number and proportion of heavy vehicles in the traffic stream will therefore result in different traffic flow conditions. This research initially outlines the different car‐following behaviour of drivers in congested heterogeneous traffic conditions indicating the necessity for developing a car‐following model, which includes these differences. A psychophysical car‐following model, similar in form to Weideman's car‐following model, was developed. Due to the complexity of the developed model, the calibration of the model was undertaken using a particle swarm optimisation algorithm with the data recorded under congested traffic conditions. This was then incorporated into a traffic microsimulation model. The results showed that the car‐following perceptual thresholds and thus action points of drivers differ based on their vehicle and the lead vehicle types. The inclusion of the heavy vehicles in the model showed significant impacts on the traffic dynamic and interactions amongst different vehicles. Copyright © 2016 John Wiley & Sons, Ltd.

Studying Non-coaxiality in Non-lane-based Car-following Behavior

“…Aghabayk et al tried to calibrate Weidman's car-following model for heavy vehicles. They found there is a significant difference between vehicles classes [19]. In another study Chen et al demonstrated that passenger car-following behavior is different in the presence of heavy vehicles [20].…”

Calibration of Gipps’ car‐following model for trucks and the impacts on fuel consumption estimation

“…The speed at the initialisation phase is defined randomly. At each iteration, the position and the speed are updated according to (3) and 4, and the archive of dominating particles is updated to include new dominants. For the MO calibration, there is not one GBEST particle because there are several objectives.…”

“…The study of car-following models calibration has already a long history, but very few studies focus specifically on heavy duty vehicles (HDVs) behaviour [1,2], although truck's reactions to traffic are different from those of cars as shown in [3][4][5]. For the truck's trajectory, one can observe that spacing would be larger, acceleration capabilities are smaller, speed profiles are more complex due to more complex engine chains.…”

“…Inertia also modifies the driving behaviour of HDVs due to the important weights of vehicles. The main studies taking into account truck following behaviour focus on the differences in following reactions [2,3,5,6] depending of the follower and the leader type (passenger car or HDV). Other existing studies which consider truck following behaviour describe the influence of heterogeneous traffic flow on traffic instability [7][8][9].…”

“…To model the heterogeneity of real road traffic, several modified car-following models were previously developed. Previous studies about truck-following rule were based either on specific mode formulation [3,8], on quite complex existing car-following rules [3] or on new developed models based on the local linear mode tree approach, LOLIMOT [10,11]. All of these studies used data from the Next Generation Simulation project.…”