多艾真体的鲁棒性归约模型  

Model of reducing robustness of multi-agent system

作　　者：龚涛[1] 蔡自兴[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410083

出　　处：《控制理论与应用》2003年第4期514-520,共7页Control Theory & Applications

基　　金：国家自然科学基金(60234030);国家博士点基金(990533177);湖南省自然科学基金(99JJY20062)

摘　　要：基于问题归约方法,对理想分布式离散型多艾真体系统鲁棒性问题进行分解和变换,消解为单艾真体的鲁棒性问题和各艾真体之间离散控制过程的鲁棒性问题,从而降低研究对象难度级别和问题元数。利用较简单鲁棒性问题的求解,就可推导出整个多艾真体系统鲁棒性判据,并总结为多艾真体鲁棒性归约模型和定理。对于实际多艾真体系统,提出多艾真体的鲁棒相关性模型,用以表示离散型多艾真体中各个艾真体之间的鲁棒相关性。然后分析多艾真体的鲁棒性归约模型与鲁棒相关性模型之间的转化关系。最后。Based on the problem reduction method, robustness problem of discrete multi-agent system in the ideal distributed environment was reduced and transformed into that of single agents and discrete control processes. So the problems were simplified and research difficulty degree was descended. Then by solving less difficult robustness problems, the robustness conditions of the whole multi-agent system could be found. Such was called the model and theorem on reducing robustness problem of multi-agent system. For real multi-agent system, the model on robustness relativity of multi-agent system was conducted advanced, the relation between the reducing model and the relativity model had been analyzed and the simulation experiment was conducted to test the models and theorem.

关 键 词：计算机控制 智能控制 鲁棒性 多艾真体 鲁棒性归约模型 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]