基于历史轨迹的城市出租车移动概率模型

对出租车的移动轨迹进行建模分析是重要的研究内容，模型需要反应节点的移动状态信息，支持用户能够对移动出租车节点进行快速查询与搜索。本文针对无法在线实时获取移动出租车实时状态信息的条件下，根据对历史轨迹信息的处理分析，提出将隐马尔科夫理论应用到城市出租车移动轨迹模型中，通过实际数据的分析建立出租车运动模型，通过对模型的计算来预测节点的位置分布概率，并在此模型上针对不同的用户需求进行查询处理，为用户提供搭车路线决策支持。通过利用真实数据集的实验证明，本文提出的方法能够较好的模拟出出租车节点的运动状态，用户也能够从模型中获取较高精度的位置状态信息。

The Moving Probability Model of Urban Cabs Based on History Trajectory

The model of cabs moving in urban is always the key point of research. The model needs reflect moving state information of cabs and much more, users could quickly query the moving cabs. In the condition of real time information of cabs is hard to obtain, we should model the moving point and forecast the state information according to moving history. This paper proposes a method which applies Hidden Markov theory to model of the moving trajectory. Though analysis of real trajectory data of San Francisco, we construct the caps moving model which is used to query the caps by users. Experiment with real datasets shows the method we proposed can simulate the moving state of caps. Users also could quickly obtain the useful location information from the model.