Safe and robust reinforcement learning for embodied intelligence with control-theoretic foundations

This exciting opportunity is based within the Mechanical and Aerospace Systems research group in the Faculty of Engineering, which conducts cutting-edge research into robotics, control, and autonomous systems with applications spanning aerospace, nuclear engineering, and embodied intelligence.

Vision

We are seeking a PhD student who is highly motivated to work at the interface of reinforcement learning, control theory, and embodied autonomous systems. You will contribute to the development of learning-based control methods that are not only high-performing but also safe, robust, and trustworthy when deployed in real-world, safety-critical environments.

Together, we will advance the foundations of intelligent autonomous systems by combining modern reinforcement learning with rigorous control-theoretic principles, enabling reliable decision-making and control in complex, uncertain, and dynamic settings.

Motivation 

Learning-enabled autonomous systems are increasingly used in applications such as unmanned aerial vehicles, robotic manipulators, and nuclear inspection and decommissioning. While reinforcement learning has demonstrated impressive performance in simulation, many existing methods lack robustness guarantees and can behave unpredictably when faced with model mismatch, uncertainty, or distribution shift.

This project is motivated by the need for deployable autonomy: learning-based systems that can operate safely over long horizons, respect physical constraints, and provide predictable closed-loop behaviour. By grounding reinforcement learning in stability theory, robust and optimal control, and physics-informed modelling, this research aims to bridge the gap between data-driven learning and dependable real-world autonomy..

Aim

You will have the opportunity to:

• Develop novel safe and robust reinforcement learning methods with strong control-theoretic foundations
• Investigate stability, robustness, and safety guarantees for learning-based control systems
• Apply and validate your methods on embodied platforms, such as UAVs, robotic systems, or simulated nuclear robotics scenarios
• Publish high-quality research in leading international journals and conferences in control, robotics, and machine learning

You will work within a multidisciplinary supervisory team spanning engineering, robotics, and computer science, and collaborate with researchers working on real-world autonomous systems.

Supervisory team

The project will be supervised by:

• Dr Anthony Siming Chen – Department of Electrical and Electronic Engineering
• Co-supervisor: Professor David Branson III - Department of Mechanical, Materials and Manufacturing Engineering
• Co-supervisor: Professor Praminda Caleb-Solly - School of Computer Science 

The supervisory team provides complementary expertise spanning control theory, reinforcement learning, robotics, and real-world autonomous systems.

Candidate requirements

We are looking for an enthusiastic, self-motivated, and resourceful candidate who is keen to pursue research at the intersection of theory and practice.

Essential requirements:

• A first-class or high 2:1 degree in a relevant subject e.g. robotics, computer science, electrical/electronic engineering, mechanical engineering, aerospace engineering, applied mathematics
• Strong interest in control theory, reinforcement learning, robotics, or autonomous systems
• Programming experience e.g. Python, MATLAB, C/C++

Desirable but not required:

• Background in control theory, dynamical systems, optimisation, or machine learning
• Experience with robotics, ROS, simulation environments, or learning-based control

Eligibility and funding

This studentship is open to UK/home and international candidates. 

After a suitable candidate is found, funding is sought from the University of Nottingham as part of a competitive process. This will cover home tuition fees and UKRI stipend.

PhD start date: October 2026

How to apply

Application deadline: 9 March 2026

If you are interested in applying, please email Dr Anthony Siming Chen at a.chen@nottingham.ac.uk 

The University of Nottingham actively supports equality, diversity and inclusion and encourages applications from all sections of society. We - the Faculty of Engineering - provide a thriving working environment for all our postgraduate researchers (PGRs) creating a strong sense of community across research disciplines. We understand that research culture is important to our PGRs so we work closely with our Postgraduate Engineering Society and PGR research group representatives to support and enhance the postgraduate research environment.

As a PGR at the University of Nottingham you will benefit from training through our Researcher Academy’s training programme. Based within the Faculty of Engineering you will have additional access to courses developed specifically for our engineering and architecture PGRs including sessions on how to write a paper, communicating your research, and research integrity. 

We offer dedicated postgraduate study spaces, have outstanding research facilities and work in partnership with leading industrial partners.