Decentralized adaptive output-feedback stabilization for large-scale stochastic nonlinear systems
Decentralized adaptive output-feedback stabilization for large-scale stochastic nonlinear systems
复制标题
DOI：
10.1016/j.automatica.2006.08.028
复制
发表时间：
2007-02-01
期刊：
影响因子：
6.4
通讯作者：
Jiang, Zhong-Ping
中科院分区：
文献类型：
--
作者：
Liu, Shu-Jun;Zhang, Ji-Feng;Jiang, Zhong-Ping
研究了具有参数不确定性、非线性不确定相互作用和随机逆动力学三种不确定性的大规模随机非线性系统的分散自适应输出反馈镇定问题。在假设子系统的逆动力学是随机输入状态稳定的前提下，采用反步法构造设计了自适应输出反馈控制器。结果表明，在某些一般条件下，闭环系统的轨迹在概率上是有界的，输出可以在概率上被调节到原点的小邻域内。此外，当漂移和扩散矢量场在原点消失时，感兴趣的平衡在概率上是全局稳定的，输出几乎可以确定地调节到原点。本文的贡献具有以下新特征：(1)即使对于集中的单输入单输出系统，本文也给出了随机、非线性、自适应、输出反馈渐近镇定的第一个结果；(2)将以前用于确定性大系统的方法推广到随机系统。同时，在随机系统设计的情况下，确定了小信号的新小增益条件；(3)允许漂移和扩散矢量场不仅依赖于可测输出，而且依赖于某些不可测状态；(4)利用参数更新规律来抵消漂移和扩散矢量场中存在的参数不确定性，这些不确定性可能出现非线性；(5)首次将随机输入状态稳定性的概念和改变供给函数的方法应用于分散控制下的随机非线性逆动力学问题。(C) 2006 Elsevier Ltd版权所有。
In this paper, the problem of decentralized adaptive output-feedback stabilization is investigated for large-scale stochastic nonlinear systems with three types of uncertainties, including parametric uncertainties, nonlinear uncertain interactions and stochastic inverse dynamics. Under the assumption that the inverse dynamics of the subsystems are stochastic input-to-state stable, an adaptive output-feedback controller is constructively designed by the backstepping method. It is shown that under some general conditions, the closed-loop system trajectories are bounded in probability and the outputs can be regulated into a small neighborhood of the origin in probability. In addition, the equilibrium of interest is globally stable in probability and the outputs can be regulated to the origin almost surely when the drift and diffusion vector fields vanish at the origin. The contributions of the work are characterized by the following novel features: (1) even for centralized single-input single-output systems, this paper presents a first result in stochastic, nonlinear, adaptive, output-feedback asymptotic stabilization; (2) the methodology previously developed for deterministic large-scale systems is generalized to stochastic ones. At the same time, novel small-gain conditions for small signals are identified in the setting of stochastic systems design; (3) both drift and diffusion vector fields are allowed to be dependent not only on the measurable outputs but some unmeasurable states; (4) parameter update laws are used to counteract the parametric uncertainty existing in both drift and diffusion vector fields, which may appear nonlinearly; (5) the concept of stochastic input-to-state stability and the method of changing supply functions are adapted, for the first time, to deal with stochastic and nonlinear inverse dynamics in the context of decentralized control. (C) 2006 Elsevier Ltd. All rights reserved.