2020-11-30 离散系统自适应控制中的一个关键性引理及证明

DR-ZF-

已于 2023-05-07 18:19:48 修改

阅读量335

点赞数

分类专栏：基础：自动控制原理\现代控制理论基础文章标签：人工智能

于 2020-11-30 22:31:11 首次发布

本文链接：https://blog.csdn.net/weixin_44382195/article/details/110393533

版权

基础：自动控制原理\现代控制理论基础专栏收录该内容

50 篇文章 31 订阅

订阅专栏

Assumption:

the expressions are all meaningful, i.e., there exists no case that division by zero.

Proof.

If ${s(t)\}$ is a bounded sequence, then by (3), $\{\|\sigma(t)\|\}$ is a bounded sequence. Then by $(1)$ and $(2)$ it follows that
$\lim _{t \rightarrow \infty} \|s(t)\|=0$

Now assume that ${\|s(t)\|\}$ is unbounded. It follows that there exists a subsequence $\left\{t_{n}\right\}$ such that
$\lim _{t_{n} \rightarrow \infty}\|s\left(t_{n}\right)\|=\infty$ and
$\|s(t)\| \leq\|s\left(t_{n}\right)\| \quad \text { for } t \leq t_{n}$
Now along the subsequence $\left\{t_{n}\right\}$
$\begin{aligned} \frac{\left[s\left(t_{n}\right)^{T}s\left(t_{n}\right)\right]^{1 / 2}}{\left[b_{1}\left(t_{n}\right)+b_{2}\left(t_{n}\right) \sigma\left(t_{n}\right)^{T} \sigma\left(t_{n}\right)\right]^{1 / 2}} & \geq \frac{\left\|s\left(t_{n}\right)\right\|}{\left[K+K \| \sigma\left(t_{n}\right)\|^{2}\right]^{1 / 2}} \\ & \geq \frac{\left\|s\left(t_{n}\right)\right\|}{K^{1 / 2}+K^{1 / 2}\| \sigma\left(t_{n}\right) \|} \\ & \geq \frac{\left\|s\left(t_{n}\right)\right\|}{{K^{1 / 2}}+K^{1 / 2}\left[C_{1}+C_{2}\left\|s\left(t_{n}\right)\right\|\right]} \end{aligned}$
Hence, when $t_n$ approaches infinity, we have
$\frac{\left[s\left(t_{n}\right)^{T}s\left(t_{n}\right)\right]^{1 / 2}}{\left[b_{1}\left(t_{n}\right)+b_{2}\left(t_{n}\right) \sigma\left(t_{n}\right)^{T} \sigma\left(t_{n}\right)\right]^{1 / 2}} \geq \frac{1}{K^{1 / 2} C_{2}}>0$
but this contradicts (1) and hence the assumption that ${\|s(t)\|\}$ is unbounded is false and the result follows.