Research Activities
The Software Engineering and Collaborative Modelling Lab is devoted to the research and technological development of new approaches to software modelling and software system production. At the present time research projects are engaged in the following areas:
  • The Distribued Computing & Software Agents Research Centre
    The development of software tools and techniques to support agent-based software interoperaability. See the links page of this site for information about available software.
  • Narrative Multimedia and Distributed Wireless Games
    Applications and technology involving the use of multimedia and narrative techniques, particularly on small mobile devices, are being developed. Work includes the following:
    • Software Cinema -- a new software engineering technique,
    • Narrative-inspired interactions mechanisms for software agents (see Klaas-Jan Winkel's thesis on this subject),
    • the FragME2004 wireless P2P platform that supports wireless game development, and
    • two wireless P2P games developed so far: RoboJoust and BOOM!.
  • Advanced Computational Architectures
    This research concerns the investigation of a novel computational architectures that can potentially provide improved problem-solving.
    • One approach has been based on the optical properties of thin-film multilayers. The architecture has been shown to have pattern-recognitions capabilities similar to neural networks. See Xiaodong Li's PhD thesis on this subject.
    • A new general approach for Evolvable Virtual Machines (EVMs) is under development that has the potential to evolve hierarhchical, cooperating agent machine structures. Some preliminary publications are available here: 1, 2, 3.
    • Adaptive knowledge discovery techniques have been investigated for improved data mining. In particular, a new market-based rule learning system (MBRL) has been developed that employs a competitive bidding process for rule agents. This has been used in combination with new techniques that were also developed for improved extraction and refinement of fuzzy rules from fuzzy neural networks. See Qing Qing Zhou's PhD thesis on this subject.
  • Process Modelling
    The modelling and simulation of complex concurrent systems is being investigated using software agents and coloured Petri nets. Examples:
    • JBees workflow system. Adaptive and distributed agent-based workflow systems for managing and monitoring the workflow in dynamic distributed systems (including systems whose structures and processes change dynamically).
    • Legal structures (legislative acts) must often cover complex interactions and so have their own behaviour. In order to understand and correctly build and refine these legal structures, reliable models of their consequential behaviour need to be constructed. An important and complex example is the New Zealand Resource Management Act of 1991, which governs the management of all natural resources in the country. See Maryam Purvis's PhD thesis and process models on this subject.
  • Computer Supported Cooperative Work and Collaborative Learning.
    This work involves the development of tools to assist the activities of collaborative groups of people that may be distributed across a network.

Related research activities:

Past projects:

  • Distributed Computing Technology & Applications
    This programme developed technology that can help organisations realise the potential of distributed interactive systems. The technical foundation of this work, software agent technology, provides mechanisms for information exchange in distributed, open systems. The research involves the ongoing development of improved agent-based development tools, supporting component services, and the demonstration of applications in the area of integrated access to distributed knowledge sources. See this page for information about available software developed under this project. (funded project completed 2007).
  • Distributed Information Systems
    The development of advanced distributed information system technology that can enable users and enterprises to access and integrate information from all available sources in a timely and useful fashion (funded project completed 2002).
  • Spatial Systems: Modelling, Analysis and Management
    The system provides analysis and simulation modules for the spatial toolbox (including feature selection, neural-network pruning, and rule extraction). Its architecture and user interface allows the modules to be distributed across a network.
  • CBIS (Connectionist-Based Information Systems)
    This was a recognised "Emerging Research Theme" at the University of Otago and focussed on developing connectionist-based algorithms and methodologies that included neural networks, fuzzy inference, evolutionary computing, distributed intelligence and multi-agent systems, and statistical learning methods.
  • Agent-Based Software Interoperability