Integrating UAV-based LiDAR and imaging data for semi-automated detection of water ponding in pavement networks

This study leverages unmanned aerial vehicles (UAV) by collecting LiDAR data and RGB imagery to detect pavement surface irregularities prone to water ponding. Such areas can impose safety risks to road users, especially under severe weather conditions. The collected data were georeferenced and imported to an ArcGIS platform to enable surface hydrological simulation. The feasibility of the proposed approach is evaluated through three case studies in an urban setting. Superimposing the surface characteristics obtained from the UAV with the hydrological models could allow for semi-automated identification of the hotspots of hydroplaning as well as accelerated moisture damage in a timely manner, which would be otherwise impractical to detect through manual inspections. Use of UAV for large-scale data collection proved to be cost-effective and timely. This approach offers a scalable solution for urban planners and infrastructure managers aiming to enhance transportation safety.