Project Details

Abstract

Rural highway work zones present elevated safety risks due to high-speed traffic, limited visibility, and dynamic construction activity, posing dangers to drivers and workers while also threatening the long-term integrity of infrastructure projects. To address these challenges, this project develops a simulation-based framework to design and evaluate a mesh sensor network capable of real-time incident detection and alerting. The proposed system integrates low-cost sensors with edge computing and self-organizing LoRaWAN (Long Range radio-carried Wide Area Network) communication to detect vehicle intrusions, equipment-lane conflicts, and worker proximity violations. Simulations will combine microscopic traffic modeling (SUMO), construction activity simulation (ROS/Gazebo), and wireless network emulation (NS-3) to enable synchronized, multi-domain testing. The system’s detection performance, message latency, localization accuracy, and communication reliability will be assessed across varied work zone scenarios and sensor configurations. Comparative evaluation against fixed-point detection, manual reporting, and cloud-based mesh networks will quantify the benefits of edge processing and network resilience. Outputs include detection benchmarks, design guidelines, and synthetic datasets of safety-critical interactions. This research advances USDOT strategic goals by improving rural safety, enabling intelligent infrastructure transformation, and preserving critical transportation assets through early detection of hazards that could disrupt construction schedules or damage facilities. The simulation framework and findings will be shared with state DOTs to support future deployment of proactive, intelligent monitoring systems.

Project Word Files