Filed Experiment Data-HV Follow AV Experiment

While mixed traffic including both autonomous vehicles (AV) and human driven vehicles (HV) is expected to be prevailing in foreseeable future, our understanding of longitudinal characteristics of mixed traffic so far is rather limited and, in particular lacks of evidence from field experiments. To bridge this gap, we design and conduct a set of field experiments to reveal the difference of the car-following behavior between a human driver following-AV and following-HV on both constant speed traffic characteristics with discrete speeds ({10,20,…,60}km/h) and dynamic car following behavior with continuous speeds (within 0-60 km/h). We recruit ten drivers for this experiment (14 runs for each driver). The experiments collect position, speed data of tested vehicles along their complete trajectories, based on which vehicle gaps, headways, and standard deviations of vehicle speed are calculated. Quantitative methods are proposed to classify drivers based on their responses in following-AV against that following-HV with both constant speed and dynamic speed characteristics. The analyses show that different drivers exhibit different behaviors in following an AV vs. an HV: some are AV-believers, some are AV-skeptics and the others are insensitive. The results suggest that depending on the characteristics and composition of the drivers, the classical car-following behavior in pure HV traffic may need to be updated properly in modeling mixed traffic in near future. This folder contains data of both const speed experiment and dynamic speed experiment, including gaps, headways and standard deviations of vehicle speed. Full-text of this paper is available at https://www.researchgate.net/publication/329772815_Field_Experiments_on_Longitudinal_Characteristics_of_Human_Driver_Behavior_Following_an_Autonomous_Vehicle