Autonomous Artificial Intelligent Agents for Bayesian Robotics

摘要

This paper addresses the problem of working with uncertain or incomplete informa-tion in multiagent systems. When we work with real life systems (like a robotic agent) wenormally use models to "insert" our knowledge to the robot. Nevertheless, any model of a realphenomenon will always be incomplete due to the existence of unknown, hidden variables thatwill influence the phenomenon, causing the model and the phenomenon to have different be-havioral patterns. In this paper we provide a new intelligent Bayesian agent architectureoriented towards Bayesian robotics that provides a framework for developing intelligent agentapplications using Bayesian theory. This architecture allows the programmer to use a commonframework for developing agents that needs to work with uncertain or incomplete information,using one data type, join probabilistic distributions, which allows to use a common exchanginguncertainty assessments in multi-agent systems.