TY  - JOUR
PY  - 2016//
TI  - A traffic flow state transition model for urban road network based on Hidden Markov Model
JO  - Neurocomputing
A1  - Zhu, Guangyu
A1  - Song, Kang
A1  - Zhang, Peng
A1  - Wang, Li
SP  - 567
EP  - 574
VL  - 214
IS  - 
N2  - Traffic guidance and prompt information could induce the change of traffic states on road sections, and in turn the effects of these changes will be transited to their relative upstream and downstream sections, which lead to dynamic variations in traffic states of urban regional road networks. In this paper, the rule of dynamic transition in traffic state of urban road networks under the effect of traffic information is studied. Specifically, the hidden Markov model is selected to represent the dynamic transition process. Then Expectation Maximization (EM) Algorithm is presented for potential traffic state estimation. Finally, verification is carried out through simulation with Variable Message Sign (VMS) selected as the information release terminal and then a certain regional road network in Beijing is chosen as the study object. The results show that the model in this paper can describe transition process of road traffic state under the effect of VMS information, and also can be used for real-time traffic state estimation of urban road networks. The study has both theoretical and practical values in evaluation of service quality of traffic information and in making traffic dispersion and control strategies for traffic management department.<p />  <p>Language: en</p>
LA  - en
SN  - 0925-2312
UR  - http://dx.doi.org/10.1016/j.neucom.2016.06.044
ID  - ref1
ER  -