KI-04-2003-Embodied-Conversational-Agents fürs web.PMD

A multiagent system (MAS) consists of a collection of individual agents, each of which displays a certain amount of autonomy with respect to its actions and perception of a domain. Overall computation is achieved by autonomous computation within each agent and by communication among the agents. The capability of the resulting MAS is an emergent functionality that may surpass the capabilities of each individual agent. It is a widely supported assumption in the multiagent community that the development of robust and scalable software systems requires autonomous agents that can complete their objectives while situated in a dynamic and uncertain environment. To this end, these agents need to be able to engage in rich, highlevel social interactions, and operate within flexible organisational structures [3]. Organisational structures institutionalise anticipated coordination, which is especially useful for mediumand large-scale applications that require limitation of the agents’ communication behaviour. Agents acting in such structures can encapsulate the complexity of subsystems (simplifying representation and design) and modularise its functionality (providing the basis for rapid development and incremental deployment). Holonic multiagent systems provide terminology and theory for the realisation of such dynamically organising agents. They transfer modularity and recursion to the agent paradigm. In a holonic multiagent system, an agent that appears as a single entity to the outside world may in fact be composed of many sub-agents and conversely, many sub-agents may decide that it is advantageous to join into the coherent structure of a superagent and thus act as single entity. We call agents consisting of sub-agents with the same inherent structure holonic agents.