This paper presents a Hidden Markov Model (HMM) based fuzzy model of short-term traffic forecasting. The model uses three sequential phases. Firstly, the HMM is used to partition the traffic flow data according to the ordering of the calculated log-likelihood values. Then, a recursive top-down algorithm is used to generate the minimum number of rules for accurate forecasting. Finally, a gradient descent method is applied to fine-tune the model parameters. To verify the method in real situation, an experimental model is constructed to produce traffic flow forecasting of freeway in CA, USA based on actual data.
HMM Based Fuzzy Model for Short-Term Traffic Forecasting
First International Conference on Transportation Information and Safety (ICTIS) ; 2011 ; Wuhan, China
ICTIS 2011 ; 1856-1862
16.06.2011
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Fuzzy Comprehensive Assessment of Short-term Traffic Flow Forecasting
British Library Online Contents | 2005
|Short-Term Traffic Flow Forecasting Using Fuzzy Logic System Methods
Online Contents | 2008
|Hybrid Neuro-Fuzzy Application in Short-Term Freeway Traffic Volume Forecasting
British Library Conference Proceedings | 2002
|Hybrid Neuro-Fuzzy Application in Short-Term Freeway Traffic Volume Forecasting
Online Contents | 2002
|Hybrid Neuro-Fuzzy Application in Short-Term Freeway Traffic Volume Forecasting
Transportation Research Record | 2002
|