Postgraduate research project

Self-adaptive artificial intelligence at the edge

Funding
Competition funded View fees and funding
Type of degree
Doctor of Philosophy
Entry requirements
2:1 honours degree View full entry requirements
Faculty graduate school
Faculty of Engineering and Physical Sciences
Closing date

About the project

Mobile and embedded systems are becoming increasingly intelligent, supported by advances in efficient machine learning (ML) models and specialised processing cores to accelerate efficient operation. In this project, you will develop software, and possibly hardware, to explore approaches for self-adaptive artificial intelligence (AI) at the edge, supporting improved system autonomy and sustainability. 

How can we make mobile and embedded computing systems more autonomous, efficient, and sustainable? Can they use advances in on-device AI to not only improve applications, but also to be self-adapt their own operation over their lifetime?

Our devices are becoming ‘smarter’, with AI embedded into their operation. This is increasingly supported within a mobile or embedded device, rather than on the cloud, for improved performance, efficiency and security. Modern processors are increasingly complex, featuring specialised heterogeneous processing elements, but users want their devices to last longer and be more sustainable.

This project will research embedding machine learning which is inherently adaptive, reacting to application needs and environmental conditions; an area which the team have made numerous recent contributions. The project will consider how it can be designed with complementary support from hardware and software, jointly optimising performance, energy and security. It will also explore its use in enabling the system to be more autonomous and introspective, providing a ’central nervous system’ to autonomously improve operation over its lifetime.

The project will adapt to your skills, experience, and interests, but is likely to involve software and model development (either for mobile platforms in C++/Python, or embedded microcontrollers in C) and experimental analysis (prototyping using development boards), and could potentially involve hardware design (embedded systems/computer engineering). 

You will be based in the Cyber Physical Systems group in the School of Electronics and Computer Science, and we will support you to attend international conferences and develop a wide range of technical and transferrable skills.