arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 0804.0558cs.AI
收藏

基于Agent的紧急情况下的环境感知

Agent-Based Perception of an Environment in an Emergency Situation

University of Le Havre(勒阿弗尔大学)

Fahem Kebair, Frédéric Serin, Cyrille Bertelle

摘要:

我们感兴趣的问题,多智能体系统发展的风险检测和应急响应在一个不确定的和部分感知的环境。在多智能体系统内部,对当前情况的评估经过三个阶段。在第一次,情况是以动态的方式表示。第二步,包括对情况进行分析,最后,将其与其他类似的已知情况进行比较。在本文中,我们提出了一个信息建模的观察环境,我们已经应用在RoboCupRescue仿真系统。来自环境的信息根据分类法并使用语义特征进行格式化。后者的定义归功于一个领域的精细本体,并由旨在动态表示当前情况的事实代理管理。

英文摘要:

We are interested in the problem of multiagent systems development for risk detecting and emergency response in an uncertain and partially perceived environment. The evaluation of the current situation passes by three stages inside the multiagent system. In a first time, the situation is represented in a dynamic way. The second step, consists to characterise the situation and finally, it is compared with other similar known situations. In this paper, we present an information modelling of an observed environment, that we have applied on the RoboCupRescue Simulation System. Information coming from the environment are formatted according to a taxonomy and using semantic features. The latter are defined thanks to a fine ontology of the domain and are managed by factual agents that aim to represent dynamically the current situation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑