摘要本文运用命令滤波的反推设计法研究了有向图下的严格反馈型非线性多智能体系统的分布式包含控制。在设计过程中，引入补偿信号消除滤波误差造成的影响，利用 RBF神经网络对系统模型中的不确定项进行逼近，通过构造 Lyapunov 函数证明所设计的分布式包含控制器能够保证闭环系统所有信号半全局一致最终有界，并通过调节参数，使包含控制误差收敛于原点的小邻域内。 本文的贡献主要有四个方面：(1)首次将命令滤波的反推设计法运用到有向图下的非线性多智能体的包含控制中；(2)引入了补偿误差信号，降低了控制器的计算复杂度，简化了稳定性的证明，提高了包含控制的精度；(3)将个体的动态模型扩展至具有控制系数的非线性系统，研究了该类多智能体的包含控制问题；(4)成功地将所设计的包含控制器应用到双连杆柔性机械臂的多智能体系统中。 32272
毕业论文关键词 包含控制 自适应控制 非线性多智能体 神经网络
Title Distributed Containment Control of Uncertain Nonlinear Multi-agent Systems
Abstract This paper studies the distributed containment control problem of nonlinear multi-agent systems in strict-feedback form via command filtered adaptive backstepping method under a directed graph. Essentially the command filter approach involves compensated signals in the design procedure which eliminate the error effect caused by command filter. The neural networks are utilized to approximate the uncertain nonlinear items. From the Lyapunov stability theory, it is proved that the distributed controller guarantee all the signals are cooperative semi-globally uniformly ultimately bounded and containment errors converge to a small neighborhood of the origin by adjusting design parameters. The contribution of the paper are four-fold. (1)The command filter method is firstly introduced to solve the distributed containment control problem under a directed graph. (2)The compensating signal involved in the design procedure is employed to alleviate the computation burden, simplify the stability analysis and improve containment control accuracy. (3)The investigated dynamic systems are extended into more general form of nonlinear systems with control coefficients. (4)The proposed distributed containment controller is successfully applied to a group of two-link flexible joint arms.
Keywords containment adaptive control nonlinear multi-agent systems neural networks
目次
1引言.1
1．1研究背景及意义1
1．2严格反馈型非线性系统的研究现状3
1．3多智能体系统的协同控制问题及其研究现状4
1．4本文主要研究的问题6
2预备知识.8
2．1图论基础8
2．2矩阵理论9
2．3神经网络.11
2．4非线性控制理论.13
3基于命令滤波的非线性多智能体系统的包含控制14
3．1问题描述.14
3．2分布式包含控制器设计.15
3．3数值算例.22
3．4本章小结.28
4具有控制系数的非线性多智能体系统的包含控制29
4．1问题描述.29
4．2主要结果.29
4．3数值算例.32
4．4本章小结.36
5多智能体系统的包含控制在双连杆柔性机械臂中的应用37
5．1问题描述.37
5．2仿真验证.38
5．3本章小结.42
结论.43
致谢.45
参考文献46
1 引言
1．1 研究背景及意义 智能体（Agent）指与环境进行交互，并且能够独立地做出决策的个体[1]。小到神经元，大到航天器，包括我们人类在内，只要拥有“独立的思想”，就是智能体。而多智能体系统（Multi-agent System）被定义为若干个智能体通过相互耦合、相互作用，来完成特定任务的大系统[1]。近年来，随着人工智能的迅猛发展，多智能体系统的研究广泛应用于机器人、航天器和无人驾驶飞机等新兴科技领域。 多智能体系统的思想源自大自然中的奇妙现象，鱼总是聚集在最富有营养的水域，大雁排成“人”字或“一”字形迁徙，蚁群通过信息素交换信息寻找巢穴和食物源的最优路径，非洲草原上数百万的角马成群有序的横渡马拉河，如 非线性多智能体系统的分布式输出包含控制:http://www.751com.cn/zidonghua/lunwen_28778.html