H.H. Wang, C.X. Du, Z.Y. Nie

Pages: 135-144

Abstract
In order to overcome the problems of low efficiency and long time in the traditional traffic capacity prediction method, in this paper, a new adaptive prediction method based on dynamic programming is proposed. This method uses approximate dynamic programming to build short-term traffic flow prediction model, obtains approximate optimal prediction results of traffic flow, and dynamically controls the smoothing coefficient of approximate optimal prediction results of traffic flow through adaptive single exponential smoothing method, so as to improve the smoothing coefficient through real data, ensure that the smoothing coefficient in the prediction process can complete independent correction, and realize adaptive prediction of traffic capacity of road traffic flow. The experimental results show that the proposed method has the advantages of high accuracy and small error rate, and the highest prediction accuracy can reach 97%.
Keywords: dynamic planning; intersection; traffic capacity; adaptive prediction