Research of bus arrival prediction model based on GPS and SVM

Bus arrival time prediction is an important part of intelligent transport system, the prediction accuracy directly affects the overall level of the intelligent transport system. This paper mainly studies the hybrid bus arrival time prediction model, which combines the real-time prediction and support vector machine (SVM) model. And in the SVM model, three factors was chosen as reference, which is time, weather, holidays. This paper is based on the analysis of the characteristics of Yuxi intelligent transport system, and choosing the support vector machine (SVM) model as prediction model, which is adaptive, robustness, and fitting small sample prediction. At the same time, using the GPS technology, which solves the problem of the complexity of the road traffic and the singularity of historical data. The model of this paper effectively solves the problem of bus arrival time prediction.

[1]  R. Jeong,et al.  Bus arrival time prediction using artificial neural network model , 2004, Proceedings. The 7th International IEEE Conference on Intelligent Transportation Systems (IEEE Cat. No.04TH8749).

[2]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[3]  Steven I-Jy Chien,et al.  Dynamic Bus Arrival Time Prediction with Artificial Neural Networks , 2002 .

[4]  Xiaobo Liu,et al.  A Dynamic Bus‐Arrival Time Prediction Model Based on APC Data , 2004 .

[5]  Carol L Schweiger.  Real-Time Bus Arrival Information Systems , 2003 .

[6]  Donald J Dailey,et al.  A PRESCRIPTION FOR TRANSIT ARRIVAL/DEPARTURE PREDICTION USING AUTOMATIC VEHICLE LOCATION DATA , 2003 .

[7]  Jan-Ming Ho,et al.  Travel time prediction with support vector regression , 2003, Proceedings of the 2003 IEEE International Conference on Intelligent Transportation Systems.