Research on the BDI-Agent Learning Model Based on Dynamic Fuzzy Logic

This paper uses the new theory tool-dynamic fuzzy logic to study the learning process with the attribute of dynamic and fuzzy as one part of BDI-agent. According to the feature with automatic reasoning of the ontology of agent, we propose the BDI-agent architecture based on the ontology and on the basis of this, we build the two different learning model of BDI-agent in MAS: (1) self-learning; (2) external learning. The combined model and the dynamic learning arithmetic are given.