Research on expressway travel time prediction based on deep learning

Accurate and efficient prediction of road travel time plays a significant role in the application of intelligent transportation system. In order to accurately predict travel time, a new attention-based CNN-BiGRU hybrid model is proposed, which can simultaneously capture the spatial-temporal features of travel time. In this model, convolutional neural network (CNN) and bi-directional gated recurrent unit (BiGRU) are used to collect the spatial and temporal characteristics of travel time separately. The attention mechanism was used for assigning different weights according to the importance of the data to further improve the prediction accuracy of the model. The model is verified by using the charging data of Guangzhou airport south line, and the experiment shows that the model can achieve accurate travel time prediction.