Abstract:Accurate prediction of traffic flow plays an important role in maintaining traffic order and traffic safety, which is a key task in the application of intelligent transportation systems (ITS). However, the urban road network has complex dynamic spatial correlation and nonlinear temporal correlation, and achieving accurate traffic flow prediction is a highly challenging task. Traditional methods use sensors deployed on roads to construct the spatial structure of the road network and capture spatial information by graph convolution. However, they ignore that the spatial correlation between nodes is dynamically changing, and using a fixed adjacency matrix cannot reflect the real road spatial structure. To overcome these limitations, this paper proposes a new spatial-temporal deep learning model: gated fusion adaptive graph neural network (GFAGNN). GFAGNN first extracts long-term dependencies on raw data through stacking expansion causal convolution, Then the spatial features of the dynamics are learned by adaptive graph attention network and adaptive graph convolutional network respectively, Finally the fused information is passed through a lightweight channel attention to extract temporal features. The experimental results on two public data sets show that our model can effectively capture the spatiotemporal correlation in traffic flow prediction. Compared with GWNET-conv model on METR-LA dataset, the three indexes in the 60-minute task prediction improved by 2.27%,2.06% and 2.13%, respectively.

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve This paper aims to address key issues in urban road traffic flow prediction. Specifically: 1. **Background and Challenges**: - Accurate traffic flow prediction is crucial for maintaining traffic order and safety, and it is a core task in intelligent transportation system (ITS) applications. - Urban road networks have complex dynamic spatial correlations and nonlinear temporal correlations, making accurate traffic flow prediction a highly challenging task. 2. **Limitations of Existing Methods**: - Traditional methods construct the spatial structure of the road network through sensors deployed on the roads and capture spatial information through graph convolution. However, these methods overlook the fact that spatial correlations between nodes are dynamically changing, and using a fixed adjacency matrix cannot reflect the true spatial structure of the roads. - Using a fixed adjacency matrix cannot encompass real spatial dependencies, and distant nodes may exhibit similar traffic flows. 3. **Proposed New Model**: - To overcome the above limitations, this paper proposes a new spatiotemporal deep learning model: Gated Fusion Adaptive Graph Neural Network (GFAGNN). GFAGNN first extracts long-term dependencies from the raw data by stacking extended causal convolutions, then learns dynamic spatial features through an adaptive graph attention network and an adaptive graph convolution network, and finally fuses information through a lightweight channel attention mechanism to extract temporal features. 4. **Experimental Results**: - Experimental results on two public datasets (METR-LA and PEMS-BAY) show that the model can effectively capture spatiotemporal correlations in traffic flow prediction. Compared to the GWNET-conv model, the three metrics for the 60-minute prediction task on the METR-LA dataset improved by 2.27%, 2.06%, and 2.13%, respectively. In summary, this paper is mainly dedicated to improving the existing graph neural network model to better capture dynamic spatiotemporal correlations, thereby enhancing the accuracy of traffic flow prediction.

Gated Fusion Adaptive Graph Neural Network for Urban Road Traffic Flow Prediction

A Spatial-Temporal Gated Hypergraph Convolution Network for Traffic Prediction

Attention Based Graph Covolution Networks for Intelligent Traffic Flow Analysis

Spatial-Temporal Traffic Flow Prediction with Fusion Graph Convolution Network and Enhanced Gated Recurrent Units

Lane-Level Traffic Flow Prediction with Heterogeneous Data and Dynamic Graphs

Spatial-temporal multi-factor fusion graph neural network for traffic prediction

Dynamic Graph Convolutional Network with Attention Fusion for Traffic Flow Prediction

Deep Spatial-Temporal Information Fusion Dynamic Graph Convolutional Network for Traffic Flow Prediction

Spatial-Temporal Fusion Graph Neural Networks for Traffic Flow Forecasting

Make More Connections: Urban Traffic Flow Forecasting with Spatiotemporal Adaptive Gated Graph Convolution Network

Traffic flow prediction based on graph convolutional networks with a parallel attention network and stacked gate recurrent units

IGAGCN: Information geometry and attention-based spatiotemporal graph convolutional networks for traffic flow prediction

Contrastive Learning-Based Adaptive Graph Fusion Convolution Network With Residual-Enhanced Decomposition Strategy for Traffic Flow Forecasting

DMGF-Net: An Efficient Dynamic Multi-Graph Fusion Network for Traffic Prediction

Dynamic multi-scale spatial-temporal graph convolutional network for traffic flow prediction

Adaptive spatial-temporal graph attention networks for traffic flow forecasting

MFDGCN: Multi-Stage Spatio-Temporal Fusion Diffusion Graph Convolutional Network for Traffic Prediction

Latent Gaussian Processes based Graph Learning for Urban Traffic Prediction

An adaptive traffic flow prediction model based on spatiotemporal graph neural network

AGCN-T: A Traffic Flow Prediction Model for Spatial-Temporal Network Dynamics

Attention-Based Gated Recurrent Graph Convolutional Network for Short-Term Traffic Flow Forecasting