Can software tools enable improved configuration, operation and exploitation of distributed and intelligent sensors for marine industrial applications

The successful exploitation of ORE resources requires accurate resource characterisation on scales smaller than that of the extraction device. Conventional acoustic doppler profilers are limited by the assumption of homogeneity on the scale of beam separation. To counteract this limitation a novel instrument was designed with multiple single beam acoustic doppler profilers converging on a single focal point. Through a series of actuators, the system is capable of spatially relocating the aforementioned focal point of the beams. In my work on the convergent-beam acoustic doppler profiler (C-ADP) I implemented a basic level of intelligence to the system such that it could "choose" which tests to run next and adjust sensor positions to account for frame bending. Whilst these choice systems were simple they represent a jumping off point for exploring the response of an underwater system to environmental stimuli, a concept I believe will be key to the next generation of marine sensing. Time synchronisation of sub-systems was also important as it allows the multiple sensors to avoid interfering with each other during data collection. Implementing a similarly precise time synchronisation in future distributed sensor networks would be even more important, avoiding interference, syncing data timestamps and allowing new applications.

A PhD offers me the next step of my journey and the perfect opportunity to use and further my skills in software development to solve real-world problems. The prospect of applying signal processing, machine learning and/or other AI tools and methods to complex data sets is very interesting and likely will make these solutions more powerful and potentially "user friendly" (if the "kit" can do more both autonomously and reliably, the end-user doesn't have to be an expert). I would seek to build upon the work I have already done in software automation and integration of sensors with intelligent software to improve data collection in marine environments. Specifically developing software tools to enable improved configuration, operation and exploitation of distributed and intelligent sensors for marine industrial applications.

As the sensors I have the most experience with, this research would start out with acoustic doppler current profilers (ADCPs) and bring in other sensors as the project progressed. An additional advantage that ADCPs present is that they typically collect several types of environmental data (velocity, pressure, temperature), offering many options for decision making based on environmental stimuli. Research would be undertaken as part of the MSM group in the Institute of Energy Systems, ideally including work with a sensor manufacturer. The inclusion of an industrial partner would allow solutions developed under PhD research to be implemented and tested on real-world problems, and further give access to the hardware and expertise of the partner. At a high-level this research and work is about low-power smart computing delivering good data and would aim to eventually be platform agnostic.