Liu, Jiucai  ORCID: https://orcid.org/0009-0001-7056-8983, Li, Haijiang ORCID: https://orcid.org/0000-0001-6326-8133 and Wang, Da-lei
2026.
Bridge digital twin system for UAV intelligent inspection: a collaborative approach based on IFC-graph and game engine.
Zhongguo Gonglu Xuebao/China Journal of Highway and Transport
39
(1)
, pp. 173-185.
10.19721/j.cnki.1001-7372.2026.01.014
|
Preview |
PDF
- Published Version
Download (7MB) | Preview |
Abstract
A bridge digital twin system tailored for UAV-based intelligent inspection must be capable of integrating and managing incremental multimodal data to support high-fidelity modeling and decision-making in real-world scenarios. However, existing digital twin platforms suffer from fragmented spatiotemporal data management, insufficient multimodal data fusion capabilities, and a lack of task-oriented decision interaction mechanisms. As a result, they fall short of forming a truly UAV-oriented bridge digital twin system.To address these challenges, this study proposes a bridge digital twin architecture that integrates the IFC (Industry Foundation Classes) standard, knowledge graphs, and game engine technologies. The system adopts an IFC-based knowledge graph (IFC-graph) as the core data management engine to unify bridge design and construction information, UAV inspection planning semantics, and field-acquired multimodal sensing data (e.g., point clouds and images). This architecture enables an incremental semantic management framework that spans multiple spatial scales-ranging from components to substructures to regions-and supports full lifecycle evolution, while also facilitating efficient semantic-based information retrieval.For virtual simulation, the system leverages Unreal Engine to build a high-fidelity 3D bridge environment that accurately reconstructs the physical scene geometry and surrounding context. Through a bidirectional linkage mechanism with the IFC knowledge graph, it supports UAV path planning, flight strategy simulation, and iterative inspection task execution, realistically modeling complex conditions such as turning, obstacle avoidance, and collisions during flight.Building upon this integrated framework, we further propose a component-aware UAV path optimization algorithm that incorporates semantic information to improve planning flexibility and precision, thereby enabling fine-grained inspection at the component level. The system has been deployed and validated in real-world bridge scenarios, demonstrating strong scalability and practical applicability. It offers a novel solution for intelligent analysis and full-lifecycle management in bridge operation and maintenance.
| Item Type: | Article |
|---|---|
| Date Type: | Publication |
| Status: | Published |
| Schools: | Schools > Engineering |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) |
| Language other than English: | Chinese |
| ISSN: | 1001-7372 |
| Date of First Compliant Deposit: | 17 March 2026 |
| Last Modified: | 01 Jun 2026 01:30 |
| URI: | https://orca.cardiff.ac.uk/id/eprint/185820 |
Actions (repository staff only)
 |
Edit Item |
Dimensions
Dimensions