PhD Studentship: Inverse Modelling of Ocean-Surface Physics for Robust Maritime Vision

Unmanned aerial and surface platforms are increasingly deployed for maritime monitoring, environmental assessment, search-and-rescue, and pollution detection. However, water surfaces are highly dynamic, optically unpredictable environments, dominated by waves, specular reflection, cloud-dependent illumination, and variable scattering. These factors significantly degrade sensor reliability, leading to missed detections, unstable feature extraction, and reduced confidence in data interpretation. Current perception pipelines treat observations as direct ground truth, yet at sea the visual signal is a complex mixture of physical inputs. There is therefore an urgent need to decouple these influences and develop laboratory-reproducible ocean optics, enabling scalable, interpretable, and repeatable sensing over open water.

The central challenge lies in recovering reliable environmental parameters from water-surface imagery in which multiple physical factors are tightly coupled. Illumination angle, spectral composition, wave geometry, surface reflection and cloud attenuation jointly shape the visual signal, making it difficult to attribute observed features to any single cause. Moreover, the dynamic and highly transient nature of waves prevents temporal consistency, hindering conventional feature tracking or motion estimation. Strong specular highlights further obscure texture, producing information-loss that standard enhancement or de-reflection methods cannot robustly recover. Compounding this, there is no scalable source of paired datasets linking environmental conditions and optical observations, which limits supervised inverse modelling and highlights the need for controlled ocean-surface optics simulation facilities.

This project introduces a physics-informed inverse modelling framework that decouples multi-factor water-surface imaging through deep neural inference and optical priors. A controlled experimental platform will replicate ocean surface optics in a large water-tank facility, generating paired datasets across variable illumination, wave states and turbidity, enabling hybrid deep-learning and iterative optimisation solutions with clear lab-to-field transferability.

Entry requirements:
We are looking for an enthusiastic, self-motivated candidate with a 1st or high 2:1 degree in robotics, computer science, physics, mechanical engineering, electrical and electronic engineering or a related field. Candidates should have programming experience and a strong interest in artificial intelligence, robotic systems, and optics. A passion for applying technological solutions to real-world challenges is essential.

English language requirements:
Applicants must meet the minimum English language requirements. Further details are available on the International website.

Funding information:
The studentship is for 3 years and provides a minimum tax-free stipend of £ 20,780 per annum (2025/26 rate) for the duration of the studentship plus university tuition fees.

Funding will be awarded on a competitive basis and is not guaranteed; availability will depend on the outcome of the selection process and subject to final approval by the University.

The following selection criteria will be used by academic schools to help them make a decision on your application.

How to apply:
All applications should be made online. Under programme name, select Electronic, Systems and Electrical Engineering. Please quote the advertised reference number: FP-SA26-PL in your application.

Applications must include a personal statement, up-to-date curriculum vitae (CV), details of two referees (one from your highest degree qualification), certified certificates and transcripts for all completed degree programmes, and a reference to the project FP-SA26-PL. Submission of a Research Proposal is not essential but may strengthen your application. Incomplete applications received after the closing date may not be considered for interview.

Shortlisted candidates will be contacted for an interview, which are expected in February/early March 2026.

Advert information