A Logic of Believable Agents

In this paper, we describe our so-called LOBA logic (a logic of believable agents). LOBA has expanded the work of KARO (van der Hoek et al., 1998 and 2000; van Linder et al., 1996 and 1995; and Meyer, 2004) and LORA logic (Woolridge, 2000 and 2002; Binh Vu Tran et al., 2003; and Woolridge and Lomuscio, 1999) and interpreted cognitive actions which only occur in agents' brains by a three-layer model. By introducing cognitive actions, we can describe the relations among an agent's cognitive components including perception, belief, emotion, desire, goal, intention, and commitment. Specially, in LOBA, we deal with emotions as a kind of modal operators

1. Introduction

It is always an interesting problem for researchers to build the model of human cognition processes. Among theories on human cognition processes, the BDI theory (Belief, Desire and Intention) put forward by Michael Bratman has been accepted by many researchers. On the basis of Bratman's theory Anand Rao and Michael Georgeff established their BDI logic. Afterward, due to many researchers' endeavor, a family of BDI logics has come into being and merged with other logics such as dynamic logic and deontic logic[12] [13] [14] [15] [16] [17] [18].