PhD Studentship: Machine-learning for High-speed Aerial Vehicle Control

The project:

This PhD project will investigate new machine learning techniques, including physics-informed neural networks for dynamical systems, for controlling high-speed and highly-manoeuvrable aerial vehicles. Emphasis will be placed on bridging the gap between classical control and modern machine learning methods, as well as high-level path planning and low-level flight control. like A principal research aim will be to develop a physics-informed, meta-learned adaptive guidance approach, which can instantly adapt the guidance and control system to new tracking targets, atmospheric conditions, or vehicle states using only a few real-time measurements.

The work will be co-funded by a leading industrial partner in the defence sector. There is also an opportunity to conduct a work placement at the industrial partner's site.

Desirable skills and experiences:

• 2.1 or above undergraduate degree in STEM


• MATLAB, Simulink


• Flight control


• Dynamical systems


• Machine learning


• Relevant academic or industrial experience

Candidate requirements:

The successful candidate must qualify for UK home student status and is expected to successfully obtain a UK security clearance. Candidates are requested to confirm their fee status when contacting one of the supervisors. The expected start date is no later than March 2027.

Funding:

fully funded

Funding: fully funded. Standard EPSRC stipend

Contacts:

Dr Duc Nguyen: duc.nguyen@bristol.ac.uk

Dr Bahadir Kocer: b.kocer@bristol.ac.uk

Professor Mark Lowenberg: m.lowenberg@bristol.ac.uk