Speaker
Description
Extended Reality for tunnel Inspection using BIM
Authors: Nicola Rimella1, Pang-jo Chun2
1Politecnico di Torino, Department of Building, Structural and Geotechnical Engineering (DISEG), Turin, Italy
2 University of Tokyo, Department of Civil Engineering, Tokyo, Japan
Abstract:
The growing challenges posed by aging infrastructure demand innovative digital solutions to ensure effective maintenance and inspection processes. In this context, the adoption of advanced technologies such as Extended Reality (XR) plays a pivotal role in modernizing infrastructure management. XR facilitates the visualization and integration of information by overlaying data-rich models onto physical environments, thereby enhancing decision-making and operational efficiency. This article presents the development of an XR application that leverages infrastructure information models to overlay inspection data onto the real world. Additionally, the application is demonstrated in a real-world case study, showcasing its potential to streamline inspection workflows and improve data accessibility in maintenance operations.
Introduction:
Inspections are fundamental for assessing the safety and preservation status of road tunnels, granting essential information for maintenance and risk management. During preliminary inspection phases, it is crucial to perform detailed visual assessments and non-destructive tests on the structure to detect damage and potential issues. The data gathered through these processes must be meticulously documented and digitized to enable in-depth analysis, including applications of machine learning algorithms. However, the digitalization process is often slow and prone to errors, which can compromise the efficiency of data utilization.
The integration of Extended Reality (XR) applications offers a transformative solution by enabling inspection data to be directly added to digital models during the inspection process, streamlining digitalization and reducing errors. Moreover, leveraging Building Information Modeling (BIM) enhances this process by linking past inspection data to virtual objects, ensuring comprehensive access to historical information during tunnel investigations.
This article explores the current use of BIM in inspection and maintenance workflows for existing tunnels, alongside an overview of XR applications in similar contexts. In the methodology section, the development process of an XR application is detailed, demonstrating how it utilizes BIM models to superimpose inspection data onto the real-world infrastructure and collect information effectively. Finally, the results section presents the application’s deployment in a real-world case study, showcasing its potential to enhance tunnel inspection and data management processes.
Relevant papers:
The use of Building Information Modeling (BIM) and Extended Reality (XR) in tunnel inspections has significant potential in improving efficiency, accuracy, and data management.
BIM serves as a central platform that integrates geometric and semantic information about tunnel structures. Frameworks like TunGPR (Zhu et al., 2024) combine BIM with deep learning for defect detection, showcasing its capability to prioritize maintenance tasks using risk assessment matrices. Data-driven maintenance, where BIM supports predictive maintenance by organizing inspection data, allowing better prioritization of repairs, and providing comprehensive insights into the structural health of tunnel linings.
AR complements BIM by overlaying digital models onto the physical environment, enabling real-time data collection and visualization during inspections. This reduces errors associated with paper-based documentation and speeds up the quality control process.
Applications like overlaying as-built BIM models onto tunnel interiors facilitate the identification of deviations from design specifications, aiding in precise displacement detection and alignment. In real-world settings, AR applications can integrate Simultaneous Localization and Mapping (SLAM) techniques for accurate spatial positioning, enabling inspectors to map and document structural elements efficiently (Behzadan et al., 2015; Melnyk et al., 2024).
Methodology:
The development process of the extended reality (XR) application begins with the creation of a BIM model of the infrastructure, based on As-Built data and inspection reports from previous evaluations. The resulting model includes objects representing individual segments of the structure, as well as objects related to damages previously documented in the reports. These damages are appropriately coded to link the already collected information. The model is further enriched with 2D damage maps, which are associated with the segments using UV mapping.
The completed model is then imported into the Unity3D game engine, where the data embedded in the BIM model can be linked to information stored in other databases. The software also utilizes XR libraries to anchor the model to the real-world environment and overlay the relevant information. At the end of the process, an Android application is developed to enable the functionalities on smartphones and tablets.
Results and conclusions:
The developed application was tested in a real-world case study involving a road tunnel in Kochi Prefecture, Japan. The testing phase evaluated the potential of XR for overlaying information, as well as the usability of features for viewing data and adding new damage records. Support from the inspection company also enabled the identification of improvements and the development of new features to enhance the inspection process.
The results demonstrate that the application effectively streamlines data visualization and collection during tunnel inspections. These findings highlight the potential of XR and BIM integration to improve inspection workflows, offering a more efficient and accurate approach to infrastructure management.
References:
Behzadan, A. H., Dong, S., & Kamat, V. R. (2015). Augmented reality visualization: A review of civil infrastructure system applications. Advanced Engineering Informatics, 29(2), 252–267. https://doi.org/10.1016/j.aei.2015.03.005
Melnyk, O., Huymajer, M., Fenzl, D., Huemer, C., Wenighofer, R., & Mazak-Huemer, A. (2024). Augmented reality for enhanced documentation and anchor inspection reporting in conventional tunnelling. Tunnelling and Underground Space Technology, 153. https://doi.org/10.1016/j.tust.2024.106040
Zhu, H., Huang, M., & Zhang, Q. B. (2024). TunGPR: Enhancing data-driven maintenance for tunnel linings through synthetic datasets, deep learning and BIM. Tunnelling and Underground Space Technology, 145. https://doi.org/10.1016/j.tust.2023.105568
