A Study on Modeling of Route Choice Mechanism of Car Drivers as Information Processing Model

本研究的目的是分析汽车驾驶员路径选择的机制，并通过引入该机制来开发交通仿真系统。在本研究中，该机制被假设为信息加工模型，在该模型中，汽车驾驶员应该按照信息加工模型来认知和行为。在这个模型中。汽车司机主要通过他们的视力来识别外部信息，并通过路标记住路线。通过重复出行，在驾驶员的认知地图上形成路线周围区域的道路网络，以检验上述模型的有效性。我们进行了长期的重复实验，受试者被要求在他们以前没有去过的地方重复开车从起点到目的地大约三个月。在实验过程中，驾驶员的视觉信息被记录在眼标记录器上。并采用认知地图法和协议法记录驾驶员的认知机制。明确了街道环境与路径选择机制的关系。而汽车驾驶员选择路线方式的长期变化也得到了明确。随着经验的增加，汽车驾驶员往往有一些有限的路线可供选择，驾驶员根据当天和时间的街道环境选择其中一条路线，为了引入上述结果，我们建立了交通仿真模型，驾驶员可以利用二元Logit模型随着道路环境的拥堵而改变路线。我们还开发了能够处理VICS等道路交通信息系统影响的仿真模型。

The purpose of this study is to analyze the mechanism of route choice of car drivers and to develop the traffic simulation system by introducing the mechanism. In this study, the mechanism is supposed as the information processing model, in which car drivers are supposed to acknowledge and act according as the information processing models. In this model. car drivers acknowledge external information mainly by their sight and remember the route by using landmarks. By repeating trips, road networks of the area around the route are formed in the cognitive map of the driver.To test the effectiveness of the above model. we conducted the long-term repeated experiment, in which the subjects are asked to drive from origin to destination repeatedly for about three months where they have not visited before. During the experiment, sight information of the driver was recorded by "eye-mark recorder". and cognitive-map method and protocol method was used to record the drivers'acknowledgment mechanism.Thorough the experiment. the relationships between street environment and route choice mechanism was made clear. And the long-term change of the way of route choice of car drivers was also made clear. As the increase of experience, car drivers tend to have some limited alternatives of the route, which drivers select one of them according to the street environment of the day and time.To introduce above results, we made traffic simulation model, in which drivers can change route as the street environment such as congestion by utilizing binary logit model. we also developed simulation model which can deal with the effect of road traffic information systems such as VICS.