Time-based line fitting method

The invention discloses a time-based line fitting method. According to the method, firstly, position information is clustered, and then screened center track points are connected together according tothe time, and relevant clustering parameters meeting the situation requirements are figured out from the angle of speed data of the track points, so that the flexibility of line fitting is improved.The method solves a problem of line fitting based on time information, and a typical data set can be extracted, so that a trajectory containing less data is fitted, the analysis efficiency is high, the line is clear, and the clustering result is stable. 本发明是种基于时间的线路拟合方法，该方法先将位置信息聚类，然后将筛选出的中心轨迹点,依据时间将其衔接起来，从轨迹点的速度数据角度出发，推算出符合情景要求的相关聚类参数，提升线路拟合的灵活性。本发明解决了基于时间信息的线路拟合问题，并且能抽取出典型数据集，从而拟合出包含较少数据量的轨迹线，分析效率高，路线清晰，聚类结果稳定。