This project develops a next-generation Agentic AI framework for intelligent infrastructure monitoring. It integrates multi-modal sensing, data-driven structural simulation, and an LLM-powered coordination system within a unified digital twin environment. The project aims to automate anomaly detection, load inference, structural prediction, and inspection workflows to support safer, more efficient, and scalable management of critical civil infrastructure.
Aim
To establish an Agentic AI system that unifies sensing, simulation, and decision-making for proactive infrastructure monitoring through multi-modal fusion, physics-informed modelling, and LLM-driven workflow orchestration.
Objectives
- Develop a digital twin foundation integrating GIS, 3D reconstruction, BIM, and real-time sensor data.
- Advance multi-modality fusion for accurate and robust anomaly detection.
- Build data-driven inverse–forward models for real-time load inference and full-field structural response prediction.
- Create an LLM-based agent to coordinate system components and automate monitoring workflows.
- Deploy and validate the full system on real infrastructure assets.
Significance
The project delivers the first closed-loop Agentic AI framework for infrastructure monitoring, addressing gaps in sensing integration, predictive modelling, and automation. It enhances anomaly detection, accelerates structural simulation, and reduces reliance on manual inspection, enabling scalable, cost-effective, and resilient management of critical assets. The system supports national priorities in infrastructure resilience and digital transformation and has broad applicability across transportation, energy, and other civil infrastructure networks.
Ideal Candidate
The Ideal candidate will have
- Willing to enrol full time on campus, preferably in January 2026
- Completion of a degree with a substantial research component
- A strong academic record supported by relevant research outputs
- Interested in applying cutting-edge AI techniques for structural engineering
Applicants must meet the entry requirements for a Higher degree by research program.
This project is open to domestic and international applicants.
Scholarship
Two scholarships are available for PhD candidates to contribute to an ARC-funded project developing a next-generation Agentic AI framework for intelligent infrastructure monitoring. The research integrates multi-modal sensing, data-driven structural simulation, and large-language-model–based system orchestration to advance digital twins and proactive maintenance for critical assets.
This scholarship provides a living stipend of $38,400 p.a. pro rata indexed, based on full-time studies, for up to a maximum of 3 years. Tuition fees offset support is also available for a successful international candidate.
Enquires
For enquires please contact Dr Qilin Li at Qilin.li@curtin.edu.au
To apply please submit a formal Expression of Interest.