Dr Linjun Lu
- Marie Sklodowska-Curie Future Roads Fellow
About
Dr Linjun Lu is currently a Marie Skłodowska-Curie Future Roads Fellow at the University of Cambridge. His research within this fellowship is dedicated to developing a pragmatic framework aiming at evaluating and enhancing the trustworthiness of digital twins for road infrastructure inspection and maintenance. Prior to his current position, Dr Lu earned his Ph.D. in civil engineering from West Virginia University in May 2023. His doctoral research focuses on the development of a suite of machine vision and AI-based algorithms tailored for intelligent transportation systems. His research interests include structural health monitoring, intelligent sensing and construction automation, infrastructure computer vision and machine learning, and intelligent transportation systems. He is also actively contributing to the academic community by serving as a reviewer for esteemed journals such as Automation in Construction, Journal of Construction Engineering and Management, and Computer-Aided Civil and Infrastructure Engineering.
Fellowship period: 15 June 2023 – 14 June 2026
Research
Project Title: A review of trustworthy requirements for road digital twins
Abstract:
Digital Twins (DT)have been recognised as a powerful tool for whole-life management of road infrastructure in an intelligent, sustainable, and resilient manner. On the way forward, however, the road infrastructure stakeholders are still faced with the challenge of how to ensure the DTs work trustily. An untrustworthy DT would produce inaccurate information and decision-making, resulting in high financial costs, inefficient road management, and safety concerns. To address this challenge, this research analyses the current state of the DT paradigm and classifies the potential factors that will impact the trustworthiness of DTs. The analysis and classification take into consideration the functionality layers of DTs and the operational requirements in road infrastructure management. Accordingly, the practical approaches that can be adopted to resolve the identified trustworthiness issues are thoroughly reviewed and systematically integrated into a framework designed to ensure the appropriate and trustworthy use of a road DT. The proposed framework not only addresses technical aspects but also aligns with operational practices, providing stakeholders with a practical roadmap for designing and deploying trustworthy DT solutions in the road sector. Specifically, the developed trustworthy framework is underpinned by three key pillars: precision, transparency, and cybersecurity. Precision is highlighted as a primary concern, with the study illustrating the importance of accurate data collection, synchronisation, and the application of machine learning for enhancing model accuracy. Transparency is underscored through the need for explainable Artificial Intelligence (AI) and Machine Learning (ML) models, ensuring that decisions are understandable to users, thereby building trust. Cybersecurity is identified as a critical aspect, with the study addressing the vulnerabilities inherent in cloud-based systems and the necessity of safeguarding against data tampering and unauthorised access. Together, these elements form a robust framework aimed at enhancing the trustworthiness of road DTs, ultimately contributing to more effective and reliable infrastructure management. It is strongly envisioned that this developed framework will serve as valuable guidance for the design and construction of robust and trustworthy digital twins, helping mitigate risks and enhance the overall effectiveness of road infrastructure management.
Alignment with SGD: This research contributes to SDG 9: Industry, Innovation, and Infrastructure and SDG 11: Sustainable Cities and Communities.
Project TRL: The current research is positioned at a TRL 1-3, focusing on basic principle observation and experimental proof of concept.