Observation and Modeling of Variability in Car-Following Behavior

本研究旨在观察和分析受试者在不知道自己是实验的一部分时的跟车行为，并建立一个模型来解释驾驶员之间和驾驶员内部的跟车行为的随机效应，以及由人的特征、交通和道路特征、环境特征等关键因素的可变性所引起的随机效应。这是对以往经验研究的重大改进，以前的经验研究是由训练有素的专业交通工程师在封闭的测试轨道上进行的。这个项目的结果应该对更现实的微观汽车跟随规律的发展做出重大贡献，这些规律包括个体驾驶员之间和内部的变化。将使用惯性、GPS、红外传感和视频设备的组合来收集数据。领跑（控制）车辆将由一名实验者驾驶，但观察的是跟随者（驾驶人群中的随机成员）的行为。录象将用于手动确认所记录的事件，以及记录可能在统计上起有关作用的某些特征，包括人、交通和道路以及环境参数。

This study aims to observe and analyze the car-following behavior of subjects who do not know they are part of an experiment, and to develop a model that can explain the stochastic effects in car-following behavior across and within drivers, as well as those caused by variability of critical factors such as human characteristics, traffic and road characteristics, and environmental characteristics. This is a significant improvement over previous empirical studies, conducted sparingly by pairs of trained professional traffic engineers on closed test tracks. The results of this project should contribute significantly to the development of more realistic microscopic car-following laws that incorporate variation across and within individual drivers. A combination of inertial, GPS, infrared sensing, and video equipment will be used to collect the data. The lead (control) vehicle will be driven by an experimenter, but it is the behavior of the follower (random members of the driving population) that will be observed. Video will be used for manual confirmation of recorded events, as well as to record certain characteristics that might play a statistically relevant role, including human, traffic and road, and environmental parameters.