-
Integration: This WP develops a Runtime Assurance Layer by deploying lightweight anomaly detection algorithms, such as autoencoders, to flag unsafe AI decisions. It also involves the development of an Ethical
-
(Edge AI) enables deploying AI algorithms and models directly on edge devices. However, AI workloads demand high performance processing, large scale data handling, and specialized hardware accelerators
-
of hardware and control algorithms for real-world applications. The Electrical Power Group is expanding rapidly with ambitious plans. We work closely with leading industry partners across multiple sectors and
-
multi-disciplinary, conducting innovative, internationally recognised research in power electronics systems. We specialise in the design and implementation of hardware and control algorithms for real
-
to operate on the edge (i.e. close to the sensor). The project will explore whether emerging logic-based ML algorithms can be translated into smaller, faster, more energy efficient and cost-effective hardware
-
hydrodynamic and water quality responses while providing robust uncertainty quantification to support reliable decision-making. Evolutionary algorithms will be employed to efficiently explore the parameter space
-
reliable decision-making. Evolutionary algorithms will be employed to efficiently explore the parameter space and undertake sensitivity analyses. The integrated framework will be validated using analytical