Search

During the deliberation phase in practical reasoning, an intelligent agent generates a set of pursuable goals (or desires) and then selects which of them he commits to achieve (intentions). When several pursuable goals are generated, it may cause that some incompatibilities arise between them. In this work, we focus on the denition, identication and resolution of these incompatibilities. The suggested approach considers the three forms of incompatibility introduced by Castelfranchi and Paglieri, namely the terminal incompatibility, the resources incompatibility and the superuity. We characterise computationally these forms of incompatibility by means of arguments that represent the plans that allow an agent to achieve his goals. The incompatibility between goals is dened based on the conicts between their plans, which are represented by means of attacks in an argumentation framework. For the goals selection, we propose to use abstract argumentation theory by applying argumentation semantics.

An intelligent agent may in general pursue multiple goals at the same time, which leads to arise some conflict among them. In this paper, we focus on these conflicts or incompatibilities among goals. Our approach is based on the model of Castelfranchi and Paglieri, in which, three forms of incompatibility and the criteria for selection of goals are defined. We characterise computationally the different forms of incompatibility (terminal, instrumental and superfluity) and propose a way, based on abstract argumentation theory, for selecting those goals that will continue to be pursued., Authors acknowledge the support of CAPES (Coordena¸c˜ao de Aperfei¸coamento de Pessoal de N´ıvel Superior) and CNPQ (National Counsel of Technological and Scientific Development) and the partial support of CSIC Intramural projects SMA (201550E040) and AMIDA (201450E009), the Spanish MICINN projects EdeTRI (TIN2012-39348-C02-01) and MILESS (TIN2013-45039-P), and the grants 2014-SGR-118 from the Generalitat de Catalunya.