A multi-agent architecture for an evolving expert system module

Abstract

This paper presents the multi-agent architecture of the expert system module included in the acquisition system Amon-Re dedicated to an application in egyptology. The main point of our approach is that this architecture has to support the evolution of the module. Indeed, the architecture is built so that the whole agent society configuration doesn't have to change when the module is evolving. Therefore, the architecture is based on data structuring and fits a natural decomposition of domain objects. Another characteristic of our architecture is dynamism. Dynamic aspect appears through the variation of the number of agents during the solving process. The agents are responsible for the part of the solving process they are in charge of. In order to get the best result as possible, they exchange their views and constraints, are able to negotiate and selfevaluate.