【作者】 魏亚辉；

【导师】 梁进；

【作者基本信息】 上海交通大学 ， 基础数学， 2011， 硕士

【摘要】 本文对多智能体的一致性进行了研究。第一章作为绪论对多智能体的概念和应用发展过程进行了简要回顾。第二章在拓扑结构下对具有时变时滞的多智能体进一致性进行了研究,假设系统无向连通且时滞上下界未知时,将平均一致性问题转化为讨论扰动方程的渐近稳定性。利用线性矩阵不等式和李雅谱诺夫泛函对系统的时滞上下界进行讨论分析,得到系统实现平均一致性的充分条件。第三章讨论了在系统网络结构是有向连通时,带有领航者的静态有向网络拓扑的多智能体系统,考虑当其中各个动态方程中包含有参数时的一致性问题。当网络具有对称加权时,设计基于邻接信息的控制输入和自适应律,达到使多智能体的状态渐近趋近于领航者的目的。更多还原

【Abstract】 This paper is devoted to the study of average consensus problems of multi-agentsystems.In Chapter one, we introduce the concepts of multi-agent systems and look backon the development in the past decades. In Chapter two, we consider average consen-sus problems in undirected networks of multi-agents with fixed topology as well asmultiple time-varying communication delays. Suppose that the topology is connected,we discuss the average consensus of multi-agent systems with multiple time-varyingdelays by the Lyapunov functional method. The result states that the multi-agents canachieve average consensus if several feasible linear matrix inequalities are providedwhere we can determine the upper bound and lower bound of time-varying communi-cation delays. In Chapter three, the leader-?ocking consensus problem with unknownparameters is studied. When the networks with topology have symmetric weights, lin-ear controllers and adaptive laws are proposed to guarantee the states of all agents cantrack the leader’s.更多还原