Michael Davies

Accepting PhD Students

PhD projects

I am currently looking for PhD students under the following themes listed below. There is the opportunity to get university funding. As these are highly competitive, you are recommended to complete a formal application as soon as possible in order to be considered for any university funding opportunities.

Current Research Themes

Machine Imaging
Computational imaging relies on the acquisition of sensor measurements that indirectly inform about the imaged object and has a broad range of applications, from computational microscopy, medical imaging (CT, MRI, ultrasound), to sonar, radar, and seismic imaging. Current state-of-the-art methods are leveraging sophisticated machine learning (ML) solutions based on deep neural networks. However, supervised ML solutions necessitate unrealistic access to a large quantity of ground truth images.

One of the aims of this theme is to develop a foundational theoretical framework and algorithmic toolbox for learning to image with limited or no ground truth data. It will lay the foundations for a new wave of unsupervised ML-based computational imaging, with potential applications across a range of settings and imaging modalities from advanced medical imaging to robotics and autonomous systems. Unleashing the ML from ground truth data will enable the algorithms to exploit the larger quantities of unsupervised measurement data available to learn more complex and effective models leading to practical benefits of accelerated acquisitions and reduced imaging artifacts, as well as totally new imaging opportunities.

Data-Driven Computational Sensing and Imaging
Today's state-of-the-art imaging and sensing rely as much on computation as they do on sensor hardware. Furthermore, computational sensing and imaging is increasingly exploiting data-driven and machine learning solutions to enhance performance and develop novel hardware/software co-designed sensing systems. However, in critical scenarios such as medicine or defence and security it is vital that verifiable algorithmic solutions are used, which places restrictions on which machine learning approaches are admissible. Importantly, fully black box machine learning solutions should be avoided. This theme will therefore focus on the development of novel algorithmic and mathematical frameworks to exploit data and machine learning for imaging and sensing within a controlled explainable and verifiable manner. There will be a specific focus on RF and electro-optic/IR sensor modalities.

Sensor and Information Fusion
Sensor networks, sensor fusion and management techniques address key challenges in intelligence, surveillance, target acquisition, and reconnaissance (ISTAR). Opportunities in adaptive data-driven sensor tasking and resource management include adaptive sensor placement, adaptive waveform design to reflect the target reflection characteristics and channel environments, and adaptive sensor selection. Although these problems have solutions in specific use cases, this theme will consider scenarios with broader applications involving multiple heterogeneous sensors on single or multiple cooperative autonomous airborne platforms.

The solutions developed in this should be robust to dynamic and congested environments, adverse weather conditions, and mutual sensor interference. A range of algorithmic and signal processing or machine learning technologies will be considered, as well as specific technical challenges. For example, projects in this theme will consider aspects related to wide area motion imaging (WAMI), position, navigation, and timing issues (PNT); robustness to adversarial attack; sensor fusion and tracking applications; use of kernel and Monte Carlo methods; outlier-robust (and other metrics) messages in belief propagation algorithms; and scheduling in large dynamic networks. Probabilistic and Bayesian frameworks will be preferred to enable uncertainty quantification and management.