To alleviate the problem of traffic congestion, accurately estimate the future transportation mode of the roundabout, and deal with its existent problems of periodic characteristics of traffic flow and short-term high volatility of traffic flow of the roundabout as a unique scene, this paper aims at the problem of short-term traffic flow prediction at the roundabout, a short-term traffic prediction model based on Gated Recurrent Unit (GRU) is proposed in this paper. The GRU unit learns the timing characteristics in the data and enhances the study of significant features by introducing the attention mechanism, enabling the learning of ultra-long sequences. In this paper, statistical data of TMU stations (traffic monitoring units) disclosed by the Highways Authority of the United Kingdom were extracted and input into the model for training and testing. The experimental results based on the measured data show that: Compared with RNN and LSTM, the structure used in this paper shows higher prediction accuracy in traffic flow prediction tasks. The mean absolute error (MAE) of the model reaches 1.9609, the root mean square error (RMSE) reaches 4.7279, and the mean percentage error (MAPE) reaches 13.1429%. The GRU-based model predicts traffic flow and its results can guide signal control optimization for better engineering applications.
Urban Roundabout Traffic Flow Prediction Based on GRU with Gated Recurrent Unit Structure
27.10.2023
1845006 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Roundabout Traffic Flow Intersection
British Library Conference Proceedings | 2007
|Traffic flow improvements at Newbridge Roundabout
British Library Conference Proceedings | 1994
|Traffic flow improvements at Newbridge Roundabout
IET Digital Library Archive | 1994
|Roundabout Aids Traffic Control
British Library Online Contents | 1998
Prediction model for short‐term traffic flow based on a K‐means‐gated recurrent unit combination
DOAJ | 2022
|