Reinforcement learning for flexible task learning on manufacturing robots.

Natural cognition provides us a lot of food for thought in creating resilient and robust artificial intelligent systems. One of the paradigms that have been created by the machine learning community is the Reinforcement Learning paradigm.

Reinforcement learning differs from supervised learning in not needing labelled input/output pairs be presented, and in not needing sub-optimal actions to be explicitly corrected. Instead the focus is on applying rewards from the environment to learn as well as finding a balance between exploration (of uncharted territory) and exploitation (of current knowledge).

This PhD topic aims to look at how we can apply this paradigm to create robots that can flexibly learn manufacturing tasks including the computational paradigms that support this.

This work will be carried out in collaboration with the renowned University of Sheffield Advanced Manufacturing Research Centre (AMRC) including the AMRC’s £43 million state-of-the-art ‘Factory 2050’ - the UK’s first fully reconfigurable assembly and component manufacturing research facility." You will also be expected to collaborate with industry on various use cases from time to time.

This is a self-funded research project. We require applicants to have either an undergraduate honours degree (2:1) or MSc (Merit or Distinction) in a relevant science or engineering subject from a reputable institution. Prospective candidates for this project should have a background in robotics, programming and embedded systems. Full details of how to apply can be found at the following link: https://www.sheffield.ac.uk/acse/research-degrees/applyphd