A Field Observation and an Analysis of Vehicle Trajectories and CO2 Emission on Time-Space Diagram for an Arterial Road

本研究的目的是获取主干道上车辆轨迹的数据，并利用该数据集阐明微观交通模拟器的可重复性。利用设在两座建筑物较高位置的若干摄像机和一辆探测车进行了实地调查。然后，利用野外实测数据，采用半自动车辆跟踪、投影变换和卡尔曼平滑方法生成车辆轨迹；通过与探测飞行器轨迹的比较，得到了较高的精度。此外，利用详细的数据集进行了精确的微观交通模拟。将仿真结果与实际数据进行了比较，探讨了仿真的可重复性特点。

The objective of this study is to obtain data for vehicle trajectories on arterial roads and clarify the reproducibility of microscopic traffic simulators using this dataset. A field survey was conducted using several video cameras located at higher positions on two buildings and a probe vehicle. Then, vehicle trajectories were generated using semi-automatic vehicle tracking, projective transformation, and a Kalman smoothing method using the data obtained by the field survey. High accuracy was confirmed to some extent by comparison with the probe vehicle trajectories. Furthermore, a precise microscopic traffic simulation was performed using the detailed dataset. The simulation results and actual data were compared, and the feature of reproducibility of the simulation was investigated.