应急可变交通网是绿色可持续发展交通下平衡有限交通设施和多变交通流量的有效交通网络。该网络由驾驶员、车辆、道路、环境、政策、法规等多种因素参与并且各因素动态变化、因素之间相互作用，使得实现该网络中的车辆群组运动建模充满挑战。本课题结合拓扑学、交通心理学、车辆物理学以及计算机科学等领域的理论方法，探索现有工作尚未涉及科学难点的解决方法，包括：(1)拟采用符号网络对路网交叉路口进行语义建模，并基于此构建路网层次语义模型，以实现应急可变交通中可变路网的建模，为车辆群组运动建模提供有效的底层数据支撑。(2)拟采用多层过滤和回归分析等方法构建驾驶员心理认知决策过程的可计算模型，以产生各种驾驶决策，用于车辆运动模拟。(3)拟构建融合车辆自身物理属性以及驾驶员心理、认知等方面的车辆群组运动综合模型，最终实现应急可变交通网中复杂车辆群组运动的模拟，为城市应急可变交通网的设置提供科学支持。

The emergency variable traffic network is one of efficient networks, which can balance traffic facilities and traffic flows well. It plays an important role in green sustainable development. The network is an emergency and complicated system produced by the interactions and integrations among drivers, vehicles, roads, environments, policies, laws and so on, which make motion simulations of vehicles in it extraordinarily challenging. Taking the advantage of basic theoretical methods from different fields, including topology, traffic psychology, vehicle physics, computer science and so on, this proposal aims to explore the algorithms for those scientific problems which are not touched by existing work: (1) we employ the signed network to depict road intersections and then propose a hierarchical semantic model to depict road networks for variable road network modeling. (2) we use the technologies of multi-layer filtering and regression analysis to model drivers’ cognition and decision-making processes for various driving decisions generating. (3) we build an integrated microscopic model, which is based on vehicles’ own physical properties and drivers’ psychological, cognitive characters, to simulate realistic traffic in the emergency variable traffic network. Through the above research, we aim to realize motion simulations of vehicles in the emergency variable traffic network, so as to provide strong support for the setting of the emergency variable traffic network.