A Simulation Study on Influence of Automated Highway System on Traffic Flow

本研究的目的是利用微观仿真模型分析AHS （Automated Highway System）的实施对道路交通流的影响。在研究中，主要做了以下工作；首先，通过汽车运行实验收集数据。基于车辆跟随理论建立微观交通流仿真模型，并根据采集到的数据对模型参数进行估计。最后，利用仿真模型分析了AHS对交通流的影响。在实验中，使用安装了GPS的小车来收集各种数据。安装在每辆车上的GPS设备每秒都会观察速度和位置，并根据这些数据计算车头时距和反应滞后数据。使用该方法，可以获得各种交通状况和道路状况的数据。观察结果总结如下：1。平均反应滞后时间在交通阻塞时间和非交通阻塞时间是不同的。非阻塞时的平均滞后时间约为2.1秒，阻塞时的平均滞后时间约为2.7秒。高速公路的滞后时间为3.4秒，比普通街道的滞后时间长1.3秒。在此基础上，建立了微观交通流仿真模型，并用该模型进行了实例分析。案例研究的重点是自动驾驶汽车的安装率，并模拟了相应的道路交通量容量的变化。因此，超过50%的比率对容量的影响较大，而超过80%的比率对容量的影响并不大。

The purpose of this study is to analyze the influence on the road traffic flow caused by the implementation of AHS (Automated Highway System) using a micro simulation model. In the research, the following work is done ; First, the data is collected through car running experiments. A microscopic traffic flow simulation model is made based on Car Following Theory and the model parameters are estimated on the basis of the collected data. Finally, the influence of AHS on traffic flow is analyzed using the simulation model.In the experiments, a variety of data is collected using GPS installed cars. The GPS device installed to each car observes the velocity and location every second and based on those data, the headway and the reaction time-lag data can be calculated. Using this method, the data on various kind of traffic conditions and road conditions can be obtained. The summary of the observation is as follows ;1.The mean reaction time lag is different between the traffic jam hours and non-jam hours. The mean time lag is approximately 2.1 seconds in non-jam hours, and contrarily it is 2.7 seconds in jam hours.2.The time lag is 3.4 seconds on express highway which is 1.3 seconds longer than that on ordinal streets.On the basis of those data, a microscopic traffic flow simulation model is made and some case studies are performed using the model. The case studies are focused on the rate of AHS installed cars and the change of road traffic volume capacity corresponding to the rate is simulated. As a result, the rate of 50% over causes a large influence to the capacity, but the rate of 80% over does not bring the more change to the capacity.