含移動方程巨觀動態車流模式之建立

Abstract

本研究目地在發展一套新的動態巨觀車流模式，並且導入交通場與機動力的概念構建此模式。在本研究中，交通場又可分為外場與內場，前者收道路外在環境之影響，後者則跟相鄰兩車之間的互動有關，交通場的大小決定了車輛在道路上的加減速行為。機動力在本研究中代表不同駕駛習慣之參數，它將扮演一個重要的因子，且所有車輛行為都將受其影響，不管是交通場的選擇門檻或者是加減速與變換車道的行為。為了讓所有模式中的變數與參數有其物理意義，本研究在構建巨觀模式之前，先行根據交通場與機動力的概念設計一微觀車流模式，並且藉由本研究設計之模擬器校估此微觀模式中各參數以及變數之間的關係，最後再將其推導成巨觀模式。我們應用蒙地卡羅法決定模擬時車輛的速度變化。除此之外，本研究處理多車道車輛行為時加入一車輛慣性之變數，此變數可以描述駕駛者在不同車道間變換所產生的衝擊，包括對他自己以及其他駕駛的影響。

The purpose of this research is to develop a new dynamic macroscopic model. In this research, we introduce the traffic fields and mobility into this new model. The traffic field is subdivided into external field and internal field, the external field is caused by the external conditions in the roads and the internal field is caused by the interactions between the adjacent cars. We assume that all cars move fast or slowly according to the magnitude of traffic field. Mobility is a key factor to affect the behaviors of vehicles and the thresholds of traffic fields. We design a microscopic model first and calibrate the model by the result of our simulator. The simulator is designed based on CA model. We define the magnitude of acceleration and deceleration in the simulator according to Monte Carlo computing technique. After that, we derive the macroscopic model from the microscopic one. It is the reason that every variables and parameters in this research are meaningful. Besides, we define a new variable, Driver’s inertia, to describe the shock caused by the lane-changing behaviors.