Artificial Intelligence for Cyberphysical Systems (CPS)
Focus of our research
Computational electronic devices are deeply embedded in modern life. Such devices sense their situation, communicate with each other and the wider Internet, act in a goal-directed manner autonomously, while also interacting with their users. These multi-faceted systems are often referred to as cyberphysical systems (CPS) to emphasise their dual computational and physical nature.
Developing intelligence for such systems must necessarily draw from leading edge approaches to machine learning, including reinforcement learning. However, this must also be grounded in the practical aspects of developing for constrained platforms. Furthermore, human interaction must also be appropriately considered.
The AI for Cyber-Physical Systems research theme focuses on practical, instantiated approaches to researching problems around intelligence in CPS reinforcement learning for the social and humanitarian engineering aspects of CPS.
We build on our group's skill-sets in data science, electronic engineering, and machine learning to develop real-world systems for practical applications.
Our flagship projects include:
DOMUS – an EU H2020 funded project to optimise the use of car cabin comfort systems to substantially extend the range of electric vehicles.
HEED – an EPSRC funded project to better understand energy interventions for displaced populations.
EnergyRev – an EPSRC project that supports the transition to decarbonisation and digitisation of UK’s Smart Local Energy Systems.
Our Postgraduate Researchers (PGRs)
|Brandi Jo Less||Fast, detailed, accurate simulation of a thermal car-cabin using machine-learning||James Bruseyfirstname.lastname@example.org|
|Enzo Iglesias||Multimode navigation for degraded UAV operation under sensor and actuator faults||James Bruseyemail@example.com|