摘要
随着实时应用的日益广泛,越来越复杂的技术已经被应用到实时系统中.在分析已有Agent模型的基础上,提出了一种新的实时Agent模型.这种模型将Agent的审慎型行为和反应型行为结合在一起,其效率比已有的Agent模型有较大的提高.还讨论了实时Agent的决策机制,提出用感知器算法对特征进行分类和任意时间算法进行决策.
With the wide application of real-time system, more and more complex techniques have been used to develop real-time systems. Multi-Agent system paradigm seems to be an appropriate approach to be applied in this area. This paper discusses the performance measures for real-time systems in detail, and a real-time Agent architecture is put forwarded on the basis of analyzing the existing models of Agent architecture and outline the challenging issues that have to be addressed in the architecture, such as using Perceptron Approach in artificial neural network to classify the features of the tasks and applying Anytime algorithm to decision etc. This architecture incorporates the deliberative behavior with reactive behavior together to get an effective solution. Finally this paper discusses the decision mechanism of the architecture presented in the paper and points out the future research trend of real-time AI systems.
出处
《小型微型计算机系统》
CSCD
北大核心
2005年第6期1033-1036,共4页
Journal of Chinese Computer Systems
基金
211重点实验室建设项目资助
湖南省科技园入园项目资金资助
