Structural Displacement Estimation of Rail Bridges Through Millimeter-Wave Radar, Accelerometers, and Non-Dedicated Multi-Modal Sensing

HOON SOHN, JIGU LEE, SUGNDONG IL, SANGJIN LEE, TACKWOO LEE, JUNHO LEE

Abstract


Rail bridges are essential components of modern transportation networks but often operate beyond their intended service life, leading to growing concerns about structural integrity and public safety. Traditional structural health monitoring (SHM) systems can be prohibitively expensive and complex, limiting their widespread use— especially for smaller rail bridges with constrained budgets. To address these challenges, we present a novel, cost-effective radar–accelerometer fusion sensor designed to provide real-time, high-precision displacement and acceleration measurements. Our approach integrates a millimeter-wave frequency-modulated continuous-wave (FMCW) radar module with a three-axis MEMS accelerometer, enabling accurate detection of even the slightest structural movements. Specifically, our sensor achieves a root-mean-square error for displacement below 0.1 mm, which is suitable for capturing subtle deformations that may indicate underlying structural issues. The hardware leverages off-the-shelf components, including an Infineon 60 GHz radar chipset and a microcontroller equipped with wireless interfaces (2.4/5.0 GHz Wi-Fi, Bluetooth, and Gigabit Ethernet), all housed in a compact (100 × 80 × 50 mm³) enclosure. Due to its streamlined design and off-the-shelf parts, mass production costs can be kept under $1,000, making it feasible for large-scale deployments. To enhance measurement fidelity, the sensor employs automatic radar target selection, guided by initial radar and acceleration data, to overcome multiple reflections and phase ambiguity in FMCW signals. This ensures robust performance under a range of field conditions and traffic scenarios. Beyond local displacement and acceleration measurements, our framework incorporates existing fiber-optic cables— originally installed for telecommunications—to form wide-area strain gauges. By using distributed fiber-optic sensing, rail bridge operators gain a broader, continuous view of structural behavior across extensive spans without requiring further dedicated sensor installations. These data sources are synchronized with a common NTP server, enabling advanced analytics that fuse displacement, strain, and vibration information into a unified, real-time rail bridge monitoring platform. Through this multi-modal integration, our radar–accelerometer fusion sensor not only addresses critical cost and scalability concerns but also provides a transformative leap in monitoring accuracy and operational efficiency. By integrating cost-effective hardware and distributed fiber-optic sensing, rail bridge stakeholders can pursue more precise, data-driven maintenance strategies, leveraging highly accurate displacement and acceleration measurements to proactively address structural concerns and extend service life


DOI
10.12783/shm2025/37454

Full Text:

PDF

Refbacks

  • There are currently no refbacks.