公交车辆行程到站时间预测模型优化研究

Research on the optimization of prediction model for bus travel time

随着智能公交系统的发展,有效提高公交车辆到站时间预测精确率,是提升公共交通服务水平,结合智能交通(ITS)发展的关键举措。在结合公交车辆GPS运行数据的基础上,构建卡尔曼滤波与支持向量机结合的预测模型。最后,应用山东省日照市19路公交线路的公交运行数据对该模型进行校验,并评价该模型的预测精确程度,评价结果表明,对公交到站时间据本文模型有较高的预测精度。

Accurately predict bus arrival time,is to improve the public transport service level,combined with the key to the development of intelligent public transport. In combination with the bus on the basis of GPS data,the establishment of kalman filtering and test bus arrival time prediction model of support vector machine（ SVM）. Finally,the application of shandong rizhao 19 road bus lines bus run data to validate the model,to evaluate the prediction effect,the model predicted results show that the model has higher prediction accuracy.