Sustainable and Scalable AI-Driven Traffic Monitoring Using Advanced Sensor Technologies
Overview
The precise monitoring and analysis of traffic situations at urban intersections have garnered significant attention in recent years, particularly with the advent of AI technologies that promise to revolutionize urban mobility. This challenge, central to effective urban traffic management, has wide-ranging implications for improving safety, reducing congestion, and optimizing traffic flow in cities worldwide. Traditional methods such as loop detectors and CCTV-based monitoring have shown potential in managing traffic, but they face practical limitations such as poor performance in low light and adverse weather conditions, high maintenance costs, limited detection capabilities, privacy concerns, and/or inflexibility to changes in traffic patterns. In contrast, while modern sensor technologies such as LiDAR and Tunneling Magneto Resistive (TMR) sensors offer advanced capabilities, their integration into comprehensive traffic monitoring systems remains underexplored.
To address these challenges, this project will explore the use of these advanced sensors, e.g., LiDAR and TMR, and their integration with traditional cameras, to enhance traffic monitoring at urban intersections. By collecting diverse data types, these sensors will enable a detailed analysis of the interactions between road users, e.g., vehicles and pedestrians. The data will be transformed into graphs that model the dynamic environment of intersections, facilitating graph neural network (GNN)-driven decision-making in smart mobility applications such as smart traffic control, traffic flow estimation, and trajectory prediction. The project will also focus on lowering the complexity of the AI models employed in this system to enable edge decision making.
A robust solution for urban traffic management that leverages cutting-edge sensor technologies and AI methodologies, balancing technological innovation with environmental and economic considerations, will be proposed. A pilot field testing will be developed to validate system performance and optimize deployment strategies, including careful sensor selection based on the specific characteristics of each intersection. This proposal also emphasizes the importance of scalability and flexibility, aiming to develop a system that can be easily adapted and expanded to various urban harsh environments.