Spatiotemporal Characteristics of Vehicle Trajectories in a Connected Vehicle Environment—A Case of an Extra-Long Tunnel Scenario

Abstract

To provide a better understanding of spatiotemporal characteristics of vehicle trajectories in connected vehicle environment, a driving simulation study was designed and conducted with an extra-long tunnel scenario. 35 drivers were recruited to participate in the driving experiment. To evaluate the spatiotemporal characteristics of vehicles with and without a warning system, objective measures were analyzed, including a spatiotemporal diagram of the curvature of obtained data and speed adjustment behaviors. This article also evaluated the impacts of connected vehicles on the traffic capacity based on the converging pattern mining method. The results indicated that the in-vehicle human-machine interface (HMI) improved driving behavior and traffic capacity. Notably, the in-vehicle HMI helped drivers better prepare for speed adjustments when approaching the tunnel and when the vehicle in front of the study vehicle made a sudden operational change. Moreover, the system contributed to a more stable operation speed, especially near the tunnel entrance, than that without the system. The findings suggest that connected vehicle environments enable drivers to change from traditional visual stimuli response behaviors to proactive response behaviors based on psychological expectations. Besides, based on the best-converging patterns from the spatiotemporal trajectories of 35 drivers, the results revealed that the traffic capacity could be improved by 22.19% under the experimental traffic flow conditions. Moreover, the differences in the benefits of the in-vehicle HMI among individuals were found to be statistically significant.