A converse Lyapunov theorem for robust exponential stochastic stability (Q4699442)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1361613
Language Label Description Also known as
default for all languages
No label defined
    English
    A converse Lyapunov theorem for robust exponential stochastic stability
    scientific article; zbMATH DE number 1361613

      Statements

      0 references
      0 references
      11 April 2000
      0 references
      Lyapunov function
      0 references
      exponential stability
      0 references
      exponential input-to-state stability
      0 references
      uniform global asymptotic stability
      0 references
      stochastic stability
      0 references
      A converse Lyapunov theorem for robust exponential stochastic stability (English)
      0 references
      We consider a stochastic system \[ dx=f(x,u)dt+g(x,u)dw, \qquad x\in{\mathbb R}^n, \quad u\in I \subset {\mathbb R}, \] where \(u\) denotes an input and \(w\) is a Wiener process, \(I\) is a compact interval, \(f\) and \(g\) are in \(C^2\), \(f(0,u)=g(0,u)=0\) for \(u\in I\), \(|f'_x|+|g'_x|\leq C_1\), \(|f'_u|+|g'_u|\leq C_2|x|\). The unique solution \(X(t,t_0,x_0,u)\) exists and is defined for all \(t\) and almost all \(\omega \in \Omega\). The main result of the paper is as follows. It is additionally assumed that \(\partial^2f/\partial x_i\partial x_j\) and \(\partial^2g/\partial x_i\partial x_j\) are bounded and for any \(x\not= 0\) there exist \(K_x>0\) and \(u_0\in I\) such that \(g(x, u_0)g^T(x,u_0)\geq K_x \text{Id}\). Then zero is exponentially uniformly globally asymptotically stable in the \(r\)-th mean (i.e. there exist constants \(C,l> 0\) such that for every \(x_0\in \mathbb R^n\) and \(u=u(t,\omega)\) it holds that \[ {\mathbb E}|X(t,t_0,x_0,u)|^r\leq C|x_0|^r\exp{(-l(t-t_0))} \] \(\forall t\in [t_0,T)\), \(T=T(\omega)\leq \infty \)) if and only if there exists a function \(V\) from \(C^{[r]}(\mathbb R^n)\cap C^\infty (\mathbb R^n\backslash \{0\})\) (Lyapunov function) such that 1. \(c_1|x|^r\leq V(x)\leq c_2|x|^r\); 2. \(|DV(x)|\leq c_3|x|^{r-1}\) \((x\not= 0)\); 3. \(DV(x)f(x,u)+\frac{1}{2}\sum_{1\leq i,j\leq n}\frac{\partial^2V}{\partial x_i\partial x_j}(g(x,u)g^T(x,u))_{i,j}\leq -c_4|x|^r\), \(\forall (x,u)\in \mathbb R^n\times I\). The Lyapunov description of exponential input-to-state stability is also given. For the entire collection see [Zbl 0922.00034].
      0 references

      Identifiers