Traffic Flow Forecasting Algorithm based on Spatio-temporal Relationship
Ying Jie Xia, Zhou Jian-bo · 2015
Accurate traffic flow forecasting is the foundation of intelligent transportation system for urban traffic control. Focusing on the nonlinear nature and spatio-temporal relationship of traffic flow, the forecasting algorithm based on spatio-temporal relationship is proposed. Firstly,grey relational analysis is used for spatio-temporal relationship analyzing. Secondly,neural network and regression analysis are applied for temporal and spatial prediction respectively. Finally,two prediction results are combined by making the weighted average. Experiments show that the proposed method has reasonable forecasting effects with high precision.