Can underwater robots spot trouble before an undersea cable fails?
Predictive robotic inspection for resilient undersea cable infrastructure
MARBLE, Croatia · 36 months
Current inspection approaches often depend on manual scheduling and offline analysis, limiting coverage and responsiveness. This PhD will develop a framework that combines risk-aware planning for AUV fleets with real-time processing of visual and sonar data.
The doctoral candidate will:
• Investigate an autonomous planning system and predictive deployment methods that account for infrastructure risk and inspection needs.
• Develop a real-time 3D reconstruction pipeline that combines visual and sonar observations adapted to underwater scenes.
• Develop and evaluate methods to detect cable damage and potential tampering during inspection.
We seek candidates with:
• A foundation in robotics, computer vision, machine learning, or autonomous planning.
• Programming skills relevant to robotics and data processing, particularly in Python or C++.
• An interest in multimodal perception, 3D reconstruction, and decision-making under uncertainty.
• The ability to conduct independent research and collaborate with researchers from different disciplines.
Experience with underwater robotics, sonar processing, neural scene representations, robotic simulation, or multi-robot systems would be useful but is not essential
Requirements
There are a few formal requirements under the Marie Skłodowska-Curie Actions (MSCA) programme:
No PhD yet: You must not already hold a doctoral degree at the time of recruitment.
Mobility rule: You must not have lived or carried out your main activity (work or studies) in the country where you will be recruited for more than 12 months during the 36 months before recruitment. MSCA exceptions may apply.
Open to all nationalities: You can apply regardless of your nationality.
University admission: You must meet the PhD admission requirements of the university associated with your position.
How to apply
Please submit your CV, copies of your Bachelor's and Master's degrees and academic transcripts, and a research statement of no more than 800 words.