In the field of modern traffic flow forecasting, short-term traffic flow forecasting is a top priority as it can be directly applied to advanced traffic information and management systems to disseminate dynamic, real-time and effective traffic information to the public, and traffic management centers. Based on the time series characteristics of traffic volume, this paper proposes a combined traffic flow prediction model of convolutional neural network (CNN) and gated recurrented unit (GRU) network which is based on deep learning. The CNN model is used to mine the parameters of traffic flow detection, and the time series features of traffic flow are mined by GRU model to realize short-term traffic prediction. The experimental results show that the combined model has an improved fit of 8.41% over the traditional long-short memory network (LSTM) model. The result indicates the effectiveness of developed model in short-term traffic forecasting.
Short-Term Traffic Prediction Based on Deep Learning
20th COTA International Conference of Transportation Professionals ; 2020 ; Xi’an, China (Conference Cancelled)
CICTP 2020 ; 3492-3502
2020-12-09
Conference paper
Electronic Resource
English
Deep learning for short-term traffic flow prediction
Elsevier | 2017
|Deep Learning for Short-Term Traffic Flow Prediction
Online Contents | 2016
|Short-term traffic flow prediction using deep learning
SLUB | 2023
|Short-term traffic flow prediction method based on deep learning
European Patent Office | 2020
|