Abstract Predicting the traffic flow is an important thing in intelligent traffic management system. Traffic can be caused due to many reasons. The reasons vary from minor road damage or blockade to major accident. The data we receive from the road is of many different structures. To improve the traffic flow prediction accuracy a combined traffic flow prediction model based on Deep learning graph convolutional neural network and long-term memory network extracts the feature of topology structure and time structure in traffic data. The residual network optimizer for all model and reduces the gradient disappearance of currents is explosion in network degradation which finally leads to traffic flow prediction. The experimental results of the combined traffic flow prediction model are more accurate than the traditional convolutional neural network.
TIME AND TOPOLOGY STRUCTURE BASED TRAFFIC FLOW PREDICTION WITH ARTIFICIAL INTELLIGENCE AND NEURAL NETWORK
2021-12-09
Patent
Elektronische Ressource
Englisch
Short-Term Traffic Flow Prediction Based on EMD and Artificial Neural Network
British Library Conference Proceedings | 2009
|Artificial Intelligence-Based Architecture for Real-Time Traffic Flow Management
Online Contents | 1998
|Artificial Intelligence-Based Architecture for Real-Time Traffic Flow Management
Transportation Research Record | 1998
|