CBR for Dynamic Situation Assessment in an Agent-Oriented Setting

摘要

. In this paper, we describe an approach of using case-based reasoning in an agent-oriented setting. CBR is used in a real-time environment to select actions of soccer players based on previously collected experiences encoded as cases. Keywords: Case-based reasoning, artificial soccer, agentoriented programming. 1 Introduction In recent years, agent-oriented technologies have caught much attention both in research and commercial areas. To provide a testbed for developing, evaluating, and testing various agent architectures, RoboCup Federation started the Robot World Cup Initiative (RoboCup), which is an attempt to foster AI and intelligent robotics research by providing a somewhat standardized problem where wide range of technologies from AI and robotics can be integrated and examined. In particular, during the annual RoboCup championships different teams utilizing different models and methods compete with each other in the domain of soccer playing. Key issues in this domain are that...