Secrecy preserving BDI agents based on answerset programming

Abstract

We consider secrecy from the point of view of an autonomous knowledge-based and resource-bound agent with incomplete and uncertain information, situated in a multi agent system. We investigate properties of secrecy and the preservation thereof in this setting and formulate desirable properties. Based on these ideas we develop a flexible BDI-based agent model and define an instance widely based on answerset programming. We show that and how our model and instance satisfy the proposed properties. We implemented our developed extendable framework for secrecy-preserving agents based on JAVA and answerset programming. © 2013 Springer-Verlag Berlin Heidelberg.