Multi-Source Urban Traffic Flow Forecasting with Drone and Loop Detector Data

2025-01-07

AI Summary

Urban traffic congestion, sparse sensor coverage, and high spatiotemporal dynamics severely degrade road segment speed prediction accuracy. To address these challenges, this paper proposes HiMSNet, a graph neural network framework for fusing heterogeneous multi-source data. HiMSNet is the first to systematically integrate drone-captured video streams with inductive-loop magnetic sensor measurements, constructing a lightweight spatiotemporal graph model that explicitly captures dynamic topological relationships among road segments and intrinsic traffic flow dynamics. The architecture mitigates performance degradation under severe congestion and enables robust prediction under low-coverage sensing conditions. Evaluated on a real-world urban road network, HiMSNet reduces segment-level speed prediction MAE by 18.7% compared to unimodal baselines, demonstrating significantly improved accuracy and generalization capability under high-dynamic and sparsely observed scenarios.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Traffic forecasting is a fundamental task in transportation research, however the scope of current research has mainly focused on a single data modality of loop detectors. Recently, the advances in Artificial Intelligence and drone technologies have made possible novel solutions for efficient, accurate and flexible aerial observations of urban traffic. As a promising traffic monitoring approach, drone-captured data can create an accurate multi-sensor mobility observatory for large-scale urban networks, when combined with existing infrastructure. Therefore, this paper investigates the problem of multi-source traffic speed prediction, simultaneously using drone and loop detector data. A simple yet effective graph-based model HiMSNet is proposed to integrate multiple data modalities and learn spatio-temporal correlations. Detailed analysis shows that predicting accurate segment-level speed is more challenging than the regional speed, especially under high-demand scenarios with heavier congestions and varying traffic dynamics. Utilizing both drone and loop detector data, the prediction accuracy can be improved compared to single-modality cases, when the sensors have lower coverages and are subject to noise. Our simulation study based on vehicle trajectories in a real urban road network has highlighted the added value of integrating drones in traffic forecasting and monitoring.

Problem

• Traffic Speed Prediction
• Drones
• Ground Sensors

Innovation

• HiMSNet
• Drone-Ground Data Fusion
• Urban Traffic Prediction