DSTGN-ExpertNet: A Deep Spatio-Temporal Graph Neural Network for High-Precision Traffic Forecasting
Journal Title: Mechatronics and Intelligent Transportation Systems - Year 2025, Vol 4, Issue 1
Abstract
Accurate traffic prediction is essential for optimizing urban mobility and mitigating congestion. Traditional deep learning models, including Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), struggle to capture complex spatiotemporal dependencies and dynamic traffic variations across urban networks. To address these challenges, this study introduces DSTGN-ExpertNet, a novel Deep Spatio-Temporal Graph Neural Network (DSTGNN) framework that integrates Graph Neural Networks (GNNs) for spatial modeling and advanced deep learning techniques for temporal dynamics. The framework employs a Mixture of Experts (MoE) approach, where specialized expert models are dynamically assigned to distinct traffic patterns through a gating network, optimizing both prediction accuracy and interpretability. The proposed model is evaluated on large-scale real-world traffic datasets from Beijing and New York, demonstrating superior performance over conventional methods, including Spatio-Temporal Graph Convolutional Networks (ST-GCN) and attention-based models. With a mean absolute error (MAE) of 1.97 on the BikeNYC dataset and 9.70 on the TaxiBJ dataset, DSTGN-ExpertNet achieves state-of-the-art accuracy. These findings highlight the potential of GNN-based frameworks in revolutionizing traffic forecasting and intelligent transportation systems (ITS).
Authors and Affiliations
Seyyed Ahmad Edalatpanah, Javad Pourqasem
Keywords
Related Articles
- EP ID EP764544
- DOI https://doi.org/10.56578/mits040103
- Views 15
- Downloads 0
Stacked Extreme Learning Machine with Horse Herd Optimization: A Methodology for Traffic Sign Recognition in Advanced Driver Assistance Systems
In the quest for autonomous vehicle safety and road infrastructure management, traffic sign recognition (TSR) remains paramount. Recent advancements in accuracy across various benchmarks have been identified in the liter...
Design and Testing of Cooperative Motion Controller for UAV-UGV System
Unmanned ground vehicles (UGVs) and quadrotor unmanned aerial vehicles (UAVs) can work together to solve challenges like intelligent transportation, thanks to their excellent performance complements in perception, loadin...
Assessing Automatic Dependent Surveillance-Broadcast Signal Quality for Airplane Departure Using Random Forest Algorithm
This study aims to assess the safety level of the Automatic Dependent Surveillance-Broadcast (ADS-B) signal quality during airplane departures at Sultan Mahmud Badaruddin II Airport. The Aero-track application was utiliz...
An Efficient Road Image Dehazing Model Based on Entropy-Weighted Gaussian Mixture Model and Level Set Refinement for Autonomous Driving Applications
Foggy road conditions present significant challenges for road monitoring systems and autonomous driving, as conventional defogging techniques often fail to accurately recover fine details of road structures, particularly...
Influence of Electromagnet-Rail Coupling on Vertical Dynamics of EMS Maglev Trains
Active control is essential for EMS maglev trains to achieve stable suspension. Currently, the main line's suspension performs well, but in areas with low track stiffness, such as the garage, turnouts, and other lines, u...