RAE2001 logo



RA5a: Structure,environment and staffing policy

Staff entered are in bold, other UWE researchers are underlined.


This submission is based on the emergence of a strong and growing interdisciplinary grouping of researchers in the Faculty of Computing, Engineering and Mathematical Sciences. This grouping exceeded the research objectives set out in the 1996 submission and continues to build on an international reputation for computer science research. The staff in this submission are attached to the following research centres: The Centre for Complex Co-operative Systems, the Intelligent Computer Systems Centre and the Intelligent Autonomous Systems Laboratory.

In the RAE96 submission we presented our plans for research development and highlighted the need to consolidate achievements and encourage the creation of new multi-disciplinary centres of excellence and to build our international research base and establish research reputations at the international level. Since then the University has achieved significantly in these areas. In the period since the last RAE submission researchers in this submission have a combined income of £2.8m (£153k per FTE), achieved 14 PhD completions and developed a number of major international research collaborations.

Headline indicators of success include:
· The Centre for Complex Co-operative Systems is recognised as an Associate Institute of CERN (the first ‘new’ UK University to be so) and runs the largest UK Flagship Teaching Company Scheme with Motorola.

· The Intelligent Computer Systems Centre’s traffic management work has been adopted by Swedish and UK government agencies (Gottenburg, Sweden and Highways Agency in the UK) and has received significant EU funding in the period.

· The Intelligent Autonomous Systems Lab has significant collaborations with Caltech and NASA Jet Propulsion Lab and received DARPA/ONR and EPSRC funding.

Achievements since 1996 demonstrate considerable gain in research quality, especially at an international level, and development of a strong research training environment for new researchers.

1.1 Centre for Complex Co-operative Systems (Baker, Barry, Chevenier, Estrella, Kovacs, Le Goff, McClatchey, Sa, Solomonides, Varga, Yang)

This was established in 1997 by McClatchey and has since grown to six members of staff, a Visiting Professor, a Visiting Fellow, three Research Fellows and nine PhD students. The CICERO project investigated the provision of a generic ‘software bus’ together with an object-oriented database in which the description of complex control systems could be captured. CICERO was funded by CERN and industrial companies from Finland, Portugal, Sweden and Belgium and involved collaboration over two years with the Universities of Helsinki, Technical University of Budapest and Madrid. In 1996/97 CICERO was completed and transferred to industry for exploitation and is now being implemented by the SpaceBel company in Belgium for industrial control systems. The expertise gained in CICERO allowed two main research themes to be developed: distributed object and process management and intelligent data-mining.

Research collaboration with CERN continued with the CRISTAL project (1997 to date) which aims to deliver distributed workflow and data management infrastructures and support technologies, and interdisciplinary research into very large scale data and process management techniques for high energy physicists and engineers. Researchers at UWE were funded by agencies such as the Royal Society and the Royal Academy of Engineering. As a consequence of the success of CRISTAL, UWE was admitted as the first Computer Science Associate Institute at CERN and McClatchey was invited to work at CERN as a Scientific Associate. The research collaboration with CERN led to the appointment of Le Goff as a Visiting Professor and of Varga as a Visiting Fellow at UWE. In addition to income recorded in RA4, £63k was received from CERN in open competition throughout its Member States (total income in the period from CERN was £171k).

A feasibility study of the industrial applications of CRISTAL was conducted with a group of manufacturing SMEs in France (sponsored by an economic development agency, Thesame AG Annecy) and a new project started with a European Bioinformatics consortium in Portugal. UWE has invested £344K in research infrastructure to support new interdisciplinary research in bioinformatics with the Centre for Research in Biomedicine (UoA 11).

CRISTAL led to a number of associated projects. Firstly the DIAMOND project (1999) between UWE and KFKI (Research Institute for Particle and Nuclear Physics) and SzTAKI (Computer and Automation Research Institute of the Hungarian Academy of Sciences). This project is funded by the British Council and the OMFB under the British-Hungarian Science and Technology Support Programme. This project investigates distributed information integration for scientific applications. Secondly the WISDOM project (1999) between UWE, CERN and the National Centre for Computing in Pakistan to study the use of XML in large-scale data migration. Thirdly a new project into the semantic integration of complex heterogeneous databases distributed across a so-called Computational Grid. This will provide discipline-independent data access tools. UWE staff are currently assisting CERN to facilitate the construction of a Grid Test-Bed to provide access to large distributed data repositories for European scientists and engineers.

Baker developed an initial Teaching Company Scheme (TCS) with Motorola European Cellular Infrastructure Division, Swindon. He won an EPSRC CASE studentship for this work and he was also awarded a Royal Academy of Engineering Industrial Fellowship. This work led to the establishment of the UK’s largest Flagship TCS, MAST (Mobile Application of Software Technologies). MAST was started in 1998 with 14 Associates and will run for five years funded at £1.5M by Motorola and the TCD. In addition, data-mining was examined in a further TCS established by Barry with the Database Group in Bristol between 1997 and 1999.

1.2 Intelligent Computer Systems Centre (Bull, Jukes, Parmee, J Smith, R Smith)

The Intelligent Computer Systems Centre was established with the objective of transferring advanced technology into industry by collaborative research projects. The centre has five members of staff, one post-doctoral researcher and five PhD students.

Fundamental research in the Centre focuses on applying and extending evolutionary computing techniques. Bull is researching into multi-agent systems (co-evolutionary computation) and co-operative multi-agent systems. There is also work undertaken by J Smith in developing mathematical models of different ‘replacement strategies’ used to manage the working memory of evolutionary computing algorithms. Initially for stationary problems, this is now being extended to non-stationary or ‘dynamic’ optimisation and noisy problem domains. Other research has focused on test landscape design and analysis, reinforcement learning control and self-adaptation of search parameters. R Smith is researching into emergent behaviour in evolving autonomous distributed agents.

The Centre is conducting a number of collaborative projects with industry in the area of distributed systems. The EU Framework IV ESPRIT trial application project TRENDS (Traffic Engineering Network Data System) delivered a real-time road traffic information service across the Internet for Gothenburg, Sweden in 1997. The Highways Agency was invited to participate as an observer in the final project reviews and this resulted in the award of the Travel Information Highway (TIH) project which started in October 1998. This project developed a CORBA based architecture, which aims to deliver a near real-time source of data about the state of the English motorway network across the Internet. The TIH forms the information dissemination route for the Traffic Control Centre (TCC) initiative from the Highways Agency; a major PFI project over the next 10 years designed to give England a strategic road management system for the first time. The TIH project is generating further research opportunities in line with the Government white paper on developing an Integrated Transport Policy.

As a further development of the relationship with APM Ltd (now Citrix UK), the Intelligent Computer Systems Centre was successful in winning the ESPRIT project FollowMe. Building on distributed object experience gained in TRENDS and basic research, Centre staff developed a software package for the deployment of mobile personal agents. The FollowMe project focused on the development of an agent scripting language based on JavaScript and XML. The centre is actively pursuing routes to develop this technology further particularly in the transport field as a consequence of expertise and reputation gained in the TIH project.

Bull conducted research into evolutionary computing and multi-agent systems which have been developed with ETRA (Spain), through an ESPRIT project VINTAGE. The Intelligent Computer Systems Centre aims to build a framework in which a set of distributed co-operative rule based agents can evolve to learn novel control strategies. The applications areas are Urban Traffic Control and real time scheduling. The Intelligent Computer Systems Centre contributes to the EU ESPRIT Network of Excellence in Evolutionary Computing project (EvoNet) as a main academic node in the network. R Smith is a member of the Management Board and J Smith is a member of the Dynamic Optimisation Working Group. The Centre has continued to develop expertise in applying machine learning based classifier systems in the steel industry through a European Coal and Steel Confederation (ECSC) funded project developing classifiers for detecting faults in rolled steels.

There have been recent developments in smaller applied projects directly funded by UK industry. In a project with BAe Sowerby, The Intelligent Computer Systems Centre is examining Evolved Emergent Multi-Agent Behaviour and in conjunction with BT, a project called, Co-evolution in a Framework of Producers/Consumers. A recent project funded by DERA is a demonstration of Evolutionary Computation Applied to Synthetic Environments for UAV Combat Scenarios. Bull currently has an EPSRC project, in collaboration with the Centre for Transport Studies at UCL, which draws on experience from the VINTAGE project to develop Adaptive Rule-based Controllers for Road Traffic Junction Signals. He also has a project with LloydsTSB examining the use of a particular evolutionary computing technique for Data Mining and with BT examining Evolutionary Computing Techniques for Collective Agents and Robotic Systems.

Parmee was formerly Director of the Engineering Design Centre at Plymouth University. His work centres on whole system design, problem decomposition, constrained and multi-objective optimisation in collaboration with BAe Systems, Rolls Royce, Lafarge Braas and British Energy. As further evidence of collaborative research the Intelligent Computer Systems Centre and the Intelligent Autonomous Systems Laboratory are working on a project in Evolutionary Collective Robotics and with Solomonides in Consolidating Medical Informatics (CoMed) based on the results of a number of smaller UK based projects and potential developments with the Centre for Complex Co-operative Systems.

1.3 Intelligent Autonomous Systems Laboratory (Adamatzky, Carse, Holland, Melhuish, McFarland, Pipe, Winfield, Zhu)

The Lab is led by Winfield (Director) and Melhuish (Projects Director). McFarland moved from Oxford University to join the Lab, in September 2000, as Professor of Biological Robotics. Additional research staff include Senior Research Fellow Adamatzky, two Research Associates, two full-time and one part-time laboratory technicians, eight PhD and two MPhil students. One PhD student holds an EPSRC industrial CASE award and another holds an EPSRC quota studentship. The Lab also has a number of part-time MPhil and PhD students, one of whom is employed at BAe Sowerby Research Centre and holds a Royal Commission for the Exhibition of 1851 Fellowship. The Lab's physical resources include a 144m2 mobile robotics arena for conducting large-scale experiments in collective robotics, believed to be the largest in the UK, and incorporates a ceiling mounted downward looking camera together with a stop-frame vision system for recording experimental runs. The Lab is also equipped with ISDN video conferencing equipment for live link-up with international collaborators.

The Lab has a particularly strong thread of research in the field of collective robotics, with an emphasis on minimalist approaches and de-centralised control (as exemplified by the PhD completion of Melhuish, 1999 Strategies for Collective Minimalist Mobile Robots). This work has been funded from a variety of external agencies including UK industry, EPSRC, US DARPA/ONR and recently DERA. Hewlett-Packard Laboratories funded an investigation into Swarm-based Telecommunications Routing; a project that explored strategies modelled on short-lived chemical markers used by social insects for dynamic routing in networks. A project with the Royal Mail, An Investigation of Intelligent Systems for the Portering of Mail, looked at co-operating robots for moving mail through automated sorting offices and successfully demonstrated a full-size proof-of-concept prototype. BAe Sowerby Research Centre funded the development of a Multi-robot system for Studying Collective Behaviours which pioneered the application of wireless local area network technology in distributed mobile robotics. The establishment of a successful research collaboration with the Center for Neuromorphic Systems Engineering, at the California Institute of Technology led directly to the development of a collective robotics laboratory at Caltech modelled on the Intelligent Autonomous Systems Lab at UWE. The Caltech Lab is equipped with UWE-designed wireless networked mobile robots (the LinuxBots). DARPA and the US Office of Naval Research (ONR) funded an investigation into robotic Plume Tracing and Odor Localization. The aim of this project is to develop and contrast single and multi-robot strategies for robotic tracking to the source of a chemical plume. Researchers at Imperial College have now also adopted the LinuxBot as an experimental platform, and further development on the LinuxBot has led to a recent patent filing (the µLinuxBot, 1999). A recent DERA award: Governing Exploration of the potential of Minimalist Autonomous Marine Sensors extends the minimalist approach to submersible robotics.

The collective robotics research has received recent awards of BAe Sowerby funding for The Flying Flock Project: Emergent Control of Groups of Miniature Flying Robots (the aim of which is to demonstrate for the first time autonomous flocking in three dimensions) and EPSRC Industrial CASE award for The Application of Helium-filled Blimp technology and Collective Robotics to Advanced Electronic Display and Messaging Systems. The Lab has an interest in all aspects of robotic autonomy, including energetic autonomy, i.e. the possibility of robots that can actively forage for energy. This interest led to an EPSRC ROPA award for an investigation into A Multi-robot System with Energetic and Computational Autonomy: a project that is generating very considerable media interest.

The Lab has had a strong parallel thread of work in adaptive or learning control in mobile robotics, including Artificial Evolution of Fuzzy and Temporal Rule Based Systems (Carse, PhD 1997), Reinforcement Learning and Knowledge Transformation in Mobile Robots (Pipe, PhD 1997) and An Evolutionary Computing Approach to Motor Learning with an Application to Robot Manipulators (Sullivan, PhD 2000). Work in the development of artificial neural network based control systems with theoretically guaranteed stability has extended earlier work in multi-axis manipulators, to systems with closed kinematic chains in Stable Adaptive Neural Control of Systems with Closed Kinematic Chains Applied to Biologically-Inspired Walking Robots (Randall, PhD 1999). This work resulted in membership of the EU Brite-Euram thematic network on Climbing and Walking Robots (CLAWAR) and collaboration with the Royal Institute of Technology, Sweden. The appointment of Zhu, and award of EPSRC Design Procedure for Neural Network Enhanced Non-Linear Control System, has strengthened the Lab’s work in neural control. A strong thread of work in Cellular Automata and Excitable Media, and a collaboration with UWE’s Chemical and Physical Sciences department, has resulted in the recent award of an EPSRC grant for the Control of Robot Navigation by Active Non-Linear Media (£181,404). A project which aims to demonstrate for the first time the use of a chemical excitable medium for massively parallel computation to solve a complex robot navigation problem.

Users of this website should note that the information is not intended to be a complete record of all research centres in the UK

Copyright 2002 - HEFCE, SHEFC, ELWa, DEL

Last updated 17 October 2003

[ Home | About the RAE2001 | Results | Submissions | Overview reports | Panels | Guidance for panel members
| Guidance for institutions | Publications  ]