关键词: 短时交通量, 混沌预测, 相空间重构, 局部多项式拟合

Abstract: Traffic flow prediction has become a key issue in intelligent transportation system study. In this paper, a prediction model of short-time traffic flow is presented based on the chaotic time series analysis. After the phase space reconstruction using traffic flow data, a two-step optimized selection method is proposed which considers Euclidean distance and equal coefficient between neighboring point and predicted point. In addition, the prediction model is developed by local polynomial method to approximate the neighboring points. The model proposed in this paper is applied to predict the real traffic flow in Dongjiang Road, Dong Guan. Comparing the traffic flow predicting value with the measure value, the results indicate that the maximal relative error is 0.445% and the minimal one is 0.038%. Moreover, single-step ahead prediction only requires 38.52 seconds. It is proved that the proposed method can significantly improve the prediction accuracy and meet the requirement of the real-time prediction.

Key words: short-term traffic flow, chaotic prediction, phase space reconstruction, local polynomial approximation