PhD Studentship in Aeronautics: AI-Driven Inverse Modelling & Design Optimisation for Next-Gen Hypersonic Flight (AE0081)

Start Date: Between 1 August 2026 and 1 July 2027

This project aims to frame hypersonic aerodynamics as a grand inverse problem. By combining modern state-of-the-art AI (foundation models, physics-informed learning) with hard physical constraints (Navier–Stokes in spectral space) we will develop methods to super-augment experimental data via data assimilation and turn sparse wind-tunnel measurements from world-class facilities into rich, high-fidelity reconstructions of complex hypersonic flow fields. This new capability will uncover hidden flow drivers and closures for unknown physics, and ultimately allow us to design robust, manufacturable, and effective passive flow control concepts using smart materials and geometries for the next wave of hypersonic flight.

You will develop an end-to-end framework that compares and blends complementary paradigms of physics informed machine learning (such as PINNs, ODIL)—to (i) super-resolve experimental data, (ii) infer unknown parameters such as the disturbance content that seeds transition and turbulent closure for mean quantities, and (iii) optimise passive control designs. The goal is breakthrough capability: turning limited data into actionable understanding and design, at speed.

What you’ll do

• Build an AI + physics assimilation pipeline that super-augments sparse measurements.
• Compile and critically evaluate an experimental database.
• Compile assimilation approaches within one coherent, fair testbed.
• Infer unknown quantities of high-speed wall-bounded flows from data, under spectral Navier–Stokes constraints.
• Use the same models to co-design passive controls (inhomogeneous materials/geometries).
• Validate on synthetic and real experiments; publish open benchmarks and papers.

Why this is exciting

• Work at the frontier of AI for science (foundation models + physics priors).
• Direct line of sight to step-change performance in hypersonic reliability and efficiency.
• Collaborations available with leading experimental facilities at Imperial and international partners.

Training & environment

You’ll gain deep skills in hypersonic flows, AI for PDEs, data assimilation, and reproducible HPC workflows (Python/C++/PyTorch/JAX). You’ll be supported with paper writing, presentations, and conference travel in a collaborative, impact-driven lab.

Supervisors: Dr Georgios Rigas, Dr Paul Bruce and Dr Denis Sipp (ONERA)

Duration: 3.5 years.

Funding: Full tuition fees and an annual tax-free stipend of £22,780 for Home, EU and International students.

Eligibility: Due to the competitive nature of these studentships, candidates will be expected to achieve/have achieved a First class honours MEng/MSci or higher degree (or international equivalent) in: Engineering, Applied Mathematics, Physics, or a closely related field

We are also looking for a strong background in aerodynamics/CFD, applied maths, or scientific computing as well as proficiency in Python/C++. Exposure to ML or automatic differentiation is a plus. You must be curious, collaborative, and motivated to turn methods into breakthroughs.

How to apply:

• Stage 1: Submit your 2-page curriculum vitae (CV), transcripts and 300-word statement explaining your motivation for applying to this PhD Studentship to: Supervisor Review Form. Supervisors will perform a comprehensive review to long-list candidates.

Deadline: 8 January 2026

• Stage 2: Supervisors will email further instructions and an application link to long-listed candidates, inviting them to make a formal application to the PhD Studentship.

Contact: For project questions: Dr Georgios Rigas

For application queries: Lisa Kelly, PhD Administrator

Equality, Diversity and Inclusion: We are an Athena SWAN Silver Award winner, a Stonewall Diversity Champion, a Disability Confident Employer and are working in partnership with GIRES to promote respect for trans people.

Advert information