James Brusey, Professor of Computer Science

James Brusey is a Professor of Computer Science with the Centre for Data Science at Coventry University, leading on AI for Cyberphysical Systems. His current research is in Machine Learning, Reinforcement Learning, and applied wireless networked sensing. Recent industrially-focused projects involve vehicle thermal comfort (with JLR), residential building and thermal comfort (Orbit Group), particle filter algorithms for flow measurement (with TUV-NEL), a new form of linear discriminant analysis (TUV-NEL), algorithms for network packet reduction (EU-funded STARGATE project with Rolls-Royce), decision support for buildings monitoring, elderly / infirm falls and near-falls sensing.

He provides thought leadership to the virtualisation framework for the EU H2020 DOMUS project (involving Fiat (CRF), Toyota (TME), and Volvo among others) that aims to revolutionise thermal comfort systems for electric vehicles.

James Brusey received his PhD from RMIT University in 2003. He has over 15 year's experience in the IT industry, part of which was as an independent consultant. He has taught research methods to all levels of researchers around the world and teaches practical courses in wireless sensing and Internet of Things to postgraduate students. He is well published and cited, has graduated 20 PhD students, and received 25 grants with a total value of £35mil. His professorship was awarded in 2018.

Reinforcement Learning shows great promise as a method to optimise the control of real-world processes and has certainly had success in playing games (e.g., AlphaGo) and dealing with simulated environments. However, we are yet to see uptake for real world problems (e.g., autonomous cars) since translating from simulated to real (or sim2real) is a massive roadblock.

In addition, traditional control doesn't attempt to optimise human-in-the-loop processes since humans are stochastic and difficult to predict.

My work aims to resolve key problems at the intersection of RL, human-centric, and real-world systems.