Agent-Based Systems

Cooperation underpins multi-agent systems in which individual agents must interact for the overall system to function effectively. Building on our previous work, we are investigating the mental components and mechanisms that are needed for autonomous agents to cooperate effectively, with a formal framework of cooperative activity. A key factor in the construction of this framework is the view that cooperation must be motivated in each of the participants. Furthermore, when autonomous agents cooperate they expose themselves to an element of risk, arising from the autonomous nature of others. Part of our work involves developing models of cooperation that account for the important roles played by the notions of trust and motivation in managing this risk. In particular we are interested in the application of the resulting models within such domains as GRID computing and Agentcities.

Contact: Nathan Griffiths

Further information can be found on the Agent-Based Systems Research Group Web pages.