Dynamic multi-agent architecture for multimedia multimodal dialogs

There is a real need for clear and sound design specifications of dynamic architecture for multimedia multimodal dialogs. These specifications must take into account the dynamic aspect in order to satisfy quality attributes such as adaptability, maintainability and understandability. Multimodal systems featuring fusion for natural human-computer interaction involve complex intelligent architectures to handle unexpected errors and mistakes made by users. These architectures should react to events that occur simultaneously with eventual redundancy from different input media. In this paper, dynamic based intelligent agent architectures for multimedia multimodal dialog protocols are proposed Global agents are decomposed into relevant sub-agents, called components. Each sub-agent is modeled separately. Timed colored Petri nets are used to specify the structure and behavior of each agent in our architecture and to prove that this architecture satisfies the desired structural and behavioral properties. This paper describes work in progress and presents a sound idea of designing dynamic architecture using agents for multimedia applications.