The objective of the study is to predict traffic flow under unusual conditions by using a deep learning model. Conventionally, machine-learning-based traffic prediction is frequently carried out. Model learning requires large amounts of training data; however, collecting sufficient samples is a challenge in the event of traffic incidents. To address this challenge, large amounts of traffic data were generated by performing traffic simulations under various traffic incidents. These data were used as training data, and a deep learning model with graph convolution and input of traffic incident information features was proposed. Subsequently, the prediction accuracy was compared with other models such as long short-term memory, which is typically used in traffic prediction. The results demonstrated the superiority of the proposed model in representing phenomena with strong spatio-temporal dependencies, such as traffic flow, and its effectiveness in traffic prediction.
Short-term prediction of traffic flow under incident conditions using graph convolutional recurrent neural network and traffic simulation
IET Intelligent Transport Systems ; 14 , 8 ; 936-946
2020-05-29
11 pages
Article (Journal)
Electronic Resource
English
machine-learning-based traffic prediction , graph convolution , prediction accuracy , traffic flow , telecommunication traffic , model learning , deep learning model , traffic simulation , learning (artificial intelligence) , graph convolutional recurrent neural network , neural nets , traffic incident information features , incident conditions , traffic incidents , traffic data , road traffic , traffic engineering computing , unusual conditions , short-term memory , recurrent neural nets , traffic simulations , training data , term prediction
Metadata by IET is licensed under CC BY 3.0
Short-term traffic flow prediction method based on graph convolution recurrent neural network
European Patent Office | 2024
|Short-term traffic flow prediction method based on integrated graph convolutional neural network
European Patent Office | 2024
|