Simulation of Complex Environments: The Fuzzy Cognitive Agent

The world is becoming increasingly competitive by the action of liberalised national and global markets. In parallel these markets have become increasingly complex making it difficult for participants to optimise their trading actions. In response, many differing computer simulation techniques have been investigated to develop either a deeper understanding of these evolving markets or to create effective system support tools. In this paper we report our efforts to develop a novel simulation platform using fuzzy cognitive agents (FCA). Our approach encapsulates fuzzy cognitive maps (FCM) generated on the Matlab Simulink platform within commercially available agent software. We firstly present our implementation of Matlab Simulink FCMs and then show how such FCMs can be integrated within a conceptual FCA architecture. Finally we report on our efforts to realise an FCA by the integration of a Matlab Simulink based FCM with the Jack Intelligent Agent Toolkit