THOMAS: MeTHods, Techniques and Tools for Open Multi-Agent Systems
Technological evolution over recent years (Internet, www, electronic commerce, wireless connection etc.) has led to a new paradigm of “computing as interaction”. Under this paradigm, computing is something that is carried out through the communication between computational entities. In this sense, computing is an inherently social activity rather than solitary, leading to new forms of conceiving, designing, developing and managing computational systems. One example of the influence of this viewpoint is the emerging model of software as a service, for example in service-oriented architectures. The technology of agents/multiagent systems is particularly promising as a support for this new computing as interaction paradigm. Dynamic agent organisations that self-adjust in order to make the most of their current environment are increasingly important. These organisations could appear in dynamic or emerging societies of agents such as Grid domains, peer-to-peer networks, or other environments in which the agents coordinate in a dynamic way in order to offer composite services. The social factors in the organisation of multi-agent systems are also increasingly important for structuring interactions in dynamic open worlds. Within this interesting area, part of the more general field of Intelligent Systems, we can identify a number of important lines of research:
- Dynamics/Regulation: flexibility in order to permit the entry and exit of agents, evolution of the organisational structure, regulation mechanisms etc.
- Heterogeneity: different types of agents with different capabilities, runtime coordination (requiring semantic description of the capacities/services), different devices (physical device resources), different channels of communication (wireless, wireline, etc.)
The aim of the present project is to advance and contribute solutions in these areas, principally in the aspects related to organisational structures. In this sense, the following is proposed:
- to develop a multiagent system architecture that is suitable for the generation of virtual organisations in open environments, as well as a support platform that will allow these systems to be implemented.
- to design a method for developing open multiagent systems which is orientated towards the concept of organisation, and which will cover the complete life cycle of an open system, allowing the suitable management of large scale complex systems and giving specific support in order to meet the possible needs of these types of systems.
- to develop a model of agents capable of taking decisions autonomously, equipped with learning mechanisms and able to respond to events, planning and replanning within runtime.
- to develop intelligent service coordination techniques/methods within open, decentralised multiagent systems: intelligent service location (directory services, syntactic and semantic comparison techniques for services) and generation and adaptation of composed services.
- to develop mechanisms based on organisational structures and virtual organisations that optimise and regulate the coordination of services in open multiagent systems.
- to define the mechanisms that allow the interaction between agents in open distributed wireless systems in which problems of communication and security may occur.
- to develop and implement the necessary security and privacy policies.
- to develop various prototypes in order to validate the proposed architecture in collaboration with companies interested in the proposal and in the results that are obtained from the project.
You can find more information about the project at http://www.thomas-tin.org/