PSI - Issue 84

Antonio Di Pietro et al. / Procedia Structural Integrity 84 (2026) 57–64

58

1. Introduction Recent structural failures have renewed interest in the development of reliable procedures for the inspection, assessment, and monitoring of existing bridges, particularly where the infrastructure network is approaching or has exceeded its design lifetime. The Italian Guidelines (MIT–CSLLPP, 2022) provide a standardised framework for bridge classification, safety evaluation, and monitoring, emphasizing accurate geometric knowledge and condition assessment, such as the identification of defects and degradation, as key elements of risk management (Ormando et al., 2022; Ormando et al., 2024). Although several advanced techniques are currently available for performing accurate assessments of existing bridges (Saidin et al., 2022; Commander, 2019, Capozzoli et al., in press), these methods are generally localised, costly, and difficult to deploy on a large scale within a knowledge-based assessment framework. Traditional visual inspections remain fundamental but present operational and safety limitations, especially when dealing with elevated or inaccessible components or emergency scenarios (Gattulli et al., 2005). These procedures often require specialised personnel and equipment, making them time-consuming and expensive. Monitoring systems deployed on the structure can measure the bridge response to travelling loads, wind and seismic actions (Bongiovanni et al., 2021, Blaso et al. 2022), and so provide information about the structural health (Clemente et al., 2019, Buffarini et al. 2023) and detect damages (Buffarini et al., 2022). Still, the implementation of such systems, especially permanent ones, can be very complex, involving high cost and requiring constant maintenance. In recent years, Unmanned Aerial Systems (UAS) have emerged as effective tools for rapid, high-resolution data acquisition, enabling safer and more flexible operations (Kim et al., 2022; Mandirola et al., 2022). Photogrammetric approaches based on Structure-from-Motion (SfM) have been widely used to derive dense point clouds (Maboudi et al., 2025); however, their performance may degrade in the presence of low-texture surfaces, illumination variability, reflective materials, or vegetation. Within this context, Light Detection and Ranging (LiDAR) sensors mounted on UAS platforms represent a significant advance. UAS-borne LiDAR falls within the broader category of Airborne Laser Scanning (ALS), which, unlike Terrestrial Laser Scanning (TLS), captures dense 3D point clouds from multiple viewpoints without requiring manual repositioning of the instrument. ALS integrates the laser scanner with a GPS receiver and an Inertial Measurement Unit (IMU), enabling accurate reconstruction of scanned surfaces by accounting for sensor trajectory, attitude, and pulse geometry (Kaartinen et al., 2022). This operational flexibility makes UAS–LiDAR particularly suitable for bridge inspections, where complex geometries, limited accessibility, and presence of vegetation or shaded areas often challenge ground-based or image-based methods. The recent development of lightweight and high frequency LiDAR payloads allows UAS to acquire homogeneous and detailed point clouds suitable for geometric documentation, structural analysis and the development of BIM and digital-twin models. Compared to TLS, UAS– LiDAR enables rapid coverage of large areas, improved visibility of hard-to-reach components, and reduced operational constraints. LiDAR has been increasingly adopted for the structural health monitoring of bridges, with applications ranging from geometric characterisation and clearance measurements to material degradation assessment and deformation monitoring (Tian et al., 2024). Previous studies have used LiDAR to determine bridge clearances with millimetre precision and to investigate the influence of environmental and loading conditions on deformation patterns. For instance, Al Shaini and Blanco (2023) employed curvature-extraction and least-squares plane-fitting methods on LiDAR point clouds of bridge decks to identify surface irregularities such as rutting, potholes, and section loss. Automated segmentation, voxelization, surface fitting, and point-based deformation algorithms have further demonstrated the potential of LiDAR to support advanced SHM workflows; however, these methods are still under development. Building on previous UAS-based digital modelling of bridges (Di Pietro et al., 2024), this paper presents a complete UAS–LiDAR methodology for bridge surveying and digital reconstruction, focusing on the specific requirements of airborne LiDAR acquisition, calibration, and processing, including trajectory–IMU alignment and strip adjustment. The methodology is composed of two main phases (Fig. 1): the UAS-LiDAR Survey and the LiDAR Data Processing. To demonstrate the operational feasibility and accuracy of the proposed workflow, a real case study on an Italian viaduct is presented. Unlike previous UAS-based studies, which have focused primarily on photogrammetric workflows, the present work introduces a fully LiDAR-driven methodology specifically optimised for bridge environments with complex morphology and limited accessibility. The contribution of this study lies in (i) the