Full Home tuition fees and annual stipend of £20,000.
This project involves the optimisation of the processing and analysis of Global Navigation Satellite Systems (GNSS) data using artificial intelligence (AI) techniques to determine accurate measurements of the sea surface from the Mayflower Autonomous Ship (MAS: mas400.com). This is a new fully autonomous, uncrewed ocean research vessel that will be the first such ship to make a trans-Atlantic crossing. The MAS enables oceanographic research opportunities over large areas.
Data collected from high precision GNSS receivers installed on such research cruises enables the measurement of sea surface height, and hence wave heights and tides to be obtained. Previously, such experiments involved the processing and analysis of data post-mission, e.g. months later. However, the development of AI aided onboard near real-time (NRT) data processing, and quality control procedures are essential for applications such as storm surge and tsunami warning systems.
NRT GNSS sea surface height estimates depend on the accuracy of satellite orbits and clocks, duration of data used and vehicle attitude. Breaking waves and signal multipath degrade the signal quality and resulting positions, whilst quality indicators computed from GNSS data alone at the observational epoch are invariably over-optimistic. For NRT “on-the-edge” accurate positioning and quality control, AI techniques will optimise GNSS data post-processing procedures developed in previous vehicle water surface height experiments. Post-processed data will be telemetered to land to optimise subsequent estimates of sea surface height.
The project will use post-processed GNSS outputs from MAS missions to build and train the AI models, and will also employ sea surface height estimates from ocean models together with wind and sea state data. This AI approach will both enable more realistic quality indicators of the NRT GNSS-measured sea surface heights to be obtained and enable on-the-fly tuning of GNSS software input parameters, e.g. data weighting and process noise. The NRT AI-developed quality indicators and AI-adaptive running of the GNSS software will be designed for porting to the MAS.
Number of awards: 1
Start date: Not later than 1st October 2022.
Award duration: 4 years
Sponsor: EPSRC and IBM UK Ltd
Dr. Nigel Penna and Dr. Miguel Morales Maqueda (Newcastle University), James Sutton (IBM UK Ltd)
You should (be about to) have at least a 2:1 Honours degree in a numerate discipline such as Maths, Physics, Oceanography, Civil Engineering, Electrical Engineering, Surveying or Computing. Experience or an interest in computer programming and scripting will be beneficial.
Studentship is available to students with a Home fee status only (all International places have been filled).
How to apply
You must apply through the University’s online postgraduate application system, and must:
- Insert code 8040F in the programme of study section
- Select programme of study ‘PhD Civil Engineering (full time)’
- Insert studentship code CASE2112 in the studentship/partnership reference field
- Attach a CV and covering letter, quoting reference code CASE2112.
- Attach degree transcripts and certificates and, if English is not your first language, a copy of your English language qualifications.
Informal enquiries to Dr. Nigel Penna (email@example.com) are welcome and encouraged before submitting a formal application to the university