The research programme encompasses a broad portfolio of projects, from large-scale European collaborations addressing smart urban transformation to focused investigations into the mathematical foundations of granular computing. Each project contributes to the overarching goal of developing computational methods that reflect and support human reasoning processes.
Flagship European Projects
REMOURBAN: REgeneration MOdel for accelerating the smart URBAN transformation
Funded under EU Horizon 2020, REMOURBAN was a major multi-partner initiative developing and validating a sustainable urban regeneration model. The project targeted energy efficiency improvements, integrated infrastructure management, and enhanced urban mobility across demonstrator cities. The INFOHUB contribution focused on data integration and decision support components within the broader smart city framework. With a total budget of 23.8 million euros, REMOURBAN brought together partners from across Europe over a five-year period (2015-2020).
HoPE: Holistic Personal public Eco-mobility
Funded by the EU Seventh Framework Programme (FP7), HoPE addressed the challenge of integrating personal and public transport modes into a seamless, environmentally conscious mobility chain. The project developed tools and services that empower travellers to make informed choices about their journeys, balancing convenience with ecological impact. INFOHUB contributed data analytics and intelligent routing components to the platform, drawing on expertise in traffic information systems and granular data processing.
MODUM: Models for Optimising Dynamic Urban Mobility
Also funded through FP7, MODUM investigated mathematical and computational models for the optimisation of urban transport networks under dynamic conditions. The project explored how real-time data streams from sensors, mobile devices, and historical records can be fused to support adaptive traffic management strategies.
Granular Computing and Information Processing
Reasoning in Complex and Uncertain Systems: A Framework for Human-centred Information Processing
This project addresses the fundamental challenge of building computational systems that reason about uncertainty in a manner consistent with human cognitive processes. By formalising the role of information granules in abstraction, aggregation, and decision-making, the project lays theoretical groundwork for more intuitive and transparent intelligent systems.
Granular Modelling: Principles and the Design Environment
Focused on establishing a principled methodology for constructing granular models of complex systems, this project develops both the mathematical foundations and the software environment needed to design, test, and deploy granular computing solutions across diverse application domains.
Granular Computation and Hypercomputation
An investigation into the theoretical limits and possibilities of granular computing, exploring connections between information granularity and models of computation that extend beyond classical Turing machines. The project examines whether granular abstractions can provide insights into hypercomputational phenomena.
Granular Modelling in Software Engineering
This project applies granular computing principles to software engineering challenges, investigating how information granulation can improve requirements analysis, design abstraction, and the management of complexity in large-scale software systems.
Granular Art: Principles of Knowledge-based Image Understanding
Bridging computational intelligence and visual perception, this project explores how granular representations of image content can support knowledge-based interpretation of visual scenes. The work draws on rough set theory and fuzzy methods to develop image understanding systems that operate at multiple levels of abstraction.
Game-theoretic Investigation of Granular Modelling of Urban Travel Decisions
Combining game theory with granular computing, this project models the decision-making behaviour of urban travellers as a strategic interaction. By representing travel choices at appropriate levels of granularity, the approach captures the bounded rationality that characterises real-world transport decisions.
Neural Computation and Vision
Recurrent Analogue Neural Networks for Retinal Modelling
This project develops biologically inspired neural network architectures that model the signal processing functions of the retina. By incorporating recurrent connections and analogue (continuous-valued) processing, the networks capture temporal dynamics and gain modulation mechanisms observed in biological retinal circuits. A recent article on deep learning and retinal modelling explores how these biological insights translate into advances in artificial vision systems.
Functional Similarity with Morphologically Diverse Neural Networks
Investigating how neural networks with different structural configurations can exhibit functionally equivalent behaviours, this project contributes to understanding the relationship between network architecture and computational capability. The findings have implications for both neuroscience and the design of artificial neural systems.
For a complete listing of funded research grants supporting these and other projects, please visit the Research page.