Fine management of transportation has put forward higher requirements for grasping the law of traffic flow,and traffic flow prediction is an important guarantee for improving traffic control and management level.A traffic flow prediction method is proposed,which consists of two modules,namely feedback module and prediction module.Traffic flow data in Shanghai is used to verify the ef-fectiveness of the method.The results show that the average,absolute and relative errors of proposed method for all intervals are less than those of the single model based on feedback mechanism and the single or hybrid model without feedback mechanism.The proposed method can effectively improve the accuracy of short-term traffic flow prediction.
关键词
智能交通系统/城市道路/混合预测模型/反馈机制/时空关联
Key words
ITS/urban streets/hybrid prediction model/feedback mechanism/spatial and temporal correlations
维普
万方数据