Our research thus presents two aspects: (i) experimental work, and (ii) theoretical work.
Our experimental work is intended to acquire first hand experience about systems of cognitive agents, if possible in the framework of actual large complex applications, and not of small prototypes, in order to better understand what are the problems and what mechanisms should be implemented. We thus need to develop first an experimental multi-agent platform which will allow us to test the various mechanisms proposed by other research teams (e.g., communication protocols, models of interaction and cooperation, agent knowledge, transfer of knowledge among agents).
Our theoretical work aims at designing a methodology (methods for the analysis and design of development environments, allowing a casual user to build systems of cognitive agents easily, while relying on a sound theoretical model. This is absolutely necessary if one wants to be able to transfer results to the industrial world with some chance of success.
Return to main page |