About the Project
Background
Recent advances in foundation models are enabling a new generation of embodied AI systems that can perceive, reason about and interact with the physical world. Vision-Language-Action (VLA) models and large-scale robot learning have demonstrated promising capabilities. However, most current systems remain largely static after training and struggle to continually learn from interaction, adapt to new environments and embodiments, and improve from their own successes and failures.
This PhD project will investigate self-evolving embodied AI, focusing on how robots can continually develop and refine spatial intelligence through interaction with the physical world. This includes understanding objects, geometry, spatial relationships, physical dynamics and action consequences, and using this knowledge for reasoning, planning and decision-making.
The overarching goal is to develop embodied agents that do not simply execute capabilities acquired during training, but can continually learn, adapt and improve through experience.
Aims and Objectives
The project will investigate representations and learning mechanisms that enable embodied agents to accumulate spatial and physical knowledge and use it to improve their behaviour over time. Specific research questions may be developed around one or more of the following themes:
- Continual and self-improving robot learning, enabling robots to learn from demonstrations, interactions, successes and failures;
- Spatial intelligence and 3D world representations for understanding objects, geometry, dynamics, affordances and action consequences;
- World models for embodied AI, learning predictive representations that support spatial reasoning, planning and continual adaptation;
- Vision-Language-Action models, integrating multimodal foundation models with spatial representations, world models and robot policies;
- Real-to-Sim-to-Real learning, using reconstructed virtual environments to generate experience and transfer learned behaviours back to the physical world;
- Cross-embodiment learning, transferring accumulated knowledge and behaviours between humans, different robots and environments.
Methodology
The research will combine modern machine learning techniques including multimodal foundation models, computer vision, reinforcement learning, imitation learning, generative world models and 3D scene representations. Methods will be evaluated using established embodied-AI and robotics benchmarks and simulation environments, with opportunities to investigate transfer to robotic platforms where appropriate.
The exact research questions and methodology will be refined with the successful candidate, based on their interests and emerging developments in this rapidly evolving area.
Expected Outcomes and Impact
The project aims to develop new algorithms and representations that enable embodied agents to continually acquire spatial and physical knowledge, learn efficiently from experience, and generalise beyond their training conditions.
The research is expected to lead to high-quality academic publications in leading international venues in machine learning, computer vision and robotics, such as NeurIPS, ICML, ICLR, CVPR, ICCV, ICRA and CoRL. The student will also be encouraged to disseminate their research through conference presentations and, where appropriate, open-source software and research artefacts.
Beyond academic contributions, the research has potential applications in general-purpose robotics, autonomous systems and intelligent agents operating in complex real-world environments.
The project is particularly suitable for candidates interested in the intersection of machine learning, computer vision, multimodal foundation models and robotics.
Training and Support
You'll receive training and guidance in research, writing and presenting skills to support your development during your PhD. You'll also cover topics such as employability skills, research management and leadership, and graduate teaching assistant training.
In addition, York Graduate Research School works alongside the Department to offer high quality training, peer to peer support, professional development advice, and opportunities to engage others with your research.
Location
Become part of our vibrant community and contribute to inspirational and life-changing research.
You will be based in the Department of Computer Science at the University of York - an exciting and welcoming hub for innovation and collaboration with a modern and inclusive working environment. In our lakeside home on Campus East, you'll benefit from world-class laboratories and collaboration spaces.
The University of York is located a short distance from York city centre. Our historic city is consistently voted as one of the friendliest, safest and best places to live in the UK.
Find out more about student life at York
Entry Requirements
- This funded PhD opportunity is open to individuals eligible to pay tuition fees at the UK (Home) rate.
- You should hold or expect to achieve the equivalent of at least a UK upper second class degree in a relevant discipline (or equivalent).
- A strong background in machine learning, artificial intelligence, computer vision, robotics or a related area is desirable. Experience with deep learning and strong programming skills (e.g. Python/PyTorch) would be particularly beneficial.
- We are willing to consider your application if you do not fit this profile, providing you are able to demonstrate that you have sufficient computer science knowledge and experience to succeed on the programme.
- We're sorry, on this occasion this opportunity is not available to international students, or to individuals who wish to study via distance learning.
How to Apply
Please submit your application online.
Please quote the project title Self-Evolving Embodied AI: Spatial Intelligence and Continual Robot Learning in your application.
Applications are accepted throughout the year and early application is recommended. If we are impressed by your application, we will invite you to immediate interview and a decision will be made shortly afterwards.
- Supporting documents you will need to submit with your application.
- More information about the application process
Contact Us
If you have any questions about this opportunity, please email d.li@york.ac.uk.
Funding Notes
This studentship is funded by Doctoral Landscape Awards (DLA)
- You will be paid an annual living allowance
- The living allowance will be paid to you in regular instalments, and usually increases each year in line with inflation.
- The studentship will cover postgraduate research fees at the UK (Home) rate for the duration of the PhD programme.
- A Research Training Support Grant will be provided to support your research-related activities.
References
https://scholar.google.com/citations?user=RPvaE3oAAAAJ&hl=en