Graded BDI models for agent architectures based on Lukasiewicz logic and propositional dynamic logic
- 人文学院－会议论文 
In the recent past, the most influential theories with respect to agent technology are the Belief-Desire-Intention (BDI) model. This paper not only expands past achievements but also makes novel proposals. It gives a very elegant way within a uniform and flexible logical framework for the graded BDI model, that is, for reasoning about probabilities and other uncertainty models, such as degrees of desires and intentions. We choose to blend the infinite-valued Lukasiewicz Logic and Propositional Dynamic Logic to formulize the graded BDI agent model. After presenting the language and semantics for this model, we propose axioms and rules for the graded BDI logic and prove a soundness and completeness result. On the basis of dealing with composite action, we illustrate interrelations between contexts for the model. It is hoped that the present study will make contributions to represent and reason under uncertainty as well as providing a formal support for engineer complex distributed systems. 漏 Springer-Verlag Berlin Heidelberg 2012.