Communicating Rational Agents: Semantics and Verification

We present a computational semantics of communicative actions for rational agent programming languages. Three indicators are used to differentiate declarative, interrogative and imperative messages which replace the usual labels to identify socalled speech acts. We introduce a multi-agent verification logic based on the computational semantics that facilitates reasoning about communicative actions. Subsequently, this multi-agent logic is embedded into a more expressive modal logic over a run-based semantics. We relate both logics and prove expressivity results. Finally, we show how the modal logic can be used to characterize communicative actions as particular speech acts and allows to verify communication among rational agents.