多智能体系统初始状态一致性应用研究  

Application research in multi-agent system about consensus on initial state

作　　者：刘孝琪 康怀祺[1] 曾超[1] 

机构地区：[1]电子科技大学电子科学技术研究院,成都611731

出　　处：《计算机工程与应用》2014年第13期53-56,共4页Computer Engineering and Applications

基　　金：航空科学基金(No.20090580013;No.20110580002);中央高校基础研究基金(No.ZYGX2009J092)

摘　　要：针对一阶多智能体系统,研究了含有时滞和不含时滞两种情形下收敛到初始状态(认为初始状态都为0)的问题,提出了一种使多智能体系统能够收敛到初始状态的一致性协议。在此一致性协议的基础上,严格推导了该多智能体系统达到初始状态的充分性判据。结合无人机编队中的应用,给出了两种情况下多智能体系统收敛到初始状态的仿真,验证了所提出的一致性协议的正确性以及充分性判据的有效性。For the first order multi-agent system with time-delay or without time-delay, the problem converging to initial-state（regarding zero as the initial-state）is studied, and a convergence protocol to make multi-agent system converge to initial state is proposed. Based on the coherence protocol, sufficient criteria are strictly derived to make the multi-agent system achieve the initial state, Finally, combined with the application of UAV formation, the simulations in two cases with multi-agent system convergence to initial state are given, and the correctness of the consensus protocol and the validity of sufficient criteria are verified.

关 键 词：多智能体系统 一致性 初始状态 李雅普诺夫泛函 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]