scientific article; zbMATH DE number 1475059

Publication date 1999

asymptotical stability neutral type nonlinear system stabilization

Control/observation systems governed by functional-differential equations (93C23) Stabilization of systems by feedback (93D15) Asymptotic stability in control theory (93D20)

Review text

The paper deals with the stabilization and sub-optimal control of neutral type time-invariant nonlinear control systems described by \[ \begin{aligned} \dot{\mathbf x}(t) &= a_1{\mathbf x}(t)+ A_2[{\mathbf x}(t-\tau_1)-{\mathbf x}(t)]+ A_3[{\mathbf x}(t-\tau_2)-{\mathbf x}(t)]\\ &\quad +B_1{\mathbf u}(t)+ B_2[{\mathbf u}(t-\tau_3)-{\mathbf u}(t)]+ f(\cdot),\\ y(t) &= C_1{\mathbf x}(t)+ C_2[{\mathbf x}(t-\tau_4)-{\mathbf x}(t)], \end{aligned}\tag{1} \] where \({\mathbf x}\in \mathbb{R}^n\), \({\mathbf u}\in\mathbb{R}^m\), \({\mathbf y}\in \mathbb{R}^p\), \(f(\cdot)= [f_j(\cdot)]_{n\times 1}\) is a nonlinear continuous function of \({\mathbf x}(t)\), \({\mathbf x}(t-\tau_1)\), \({\mathbf x}(t-\tau_2)\) and \({\mathbf u}(t)\), the time-delays \(\tau_r\geq 0\), \(r=1,2,3,4\) are constants or bounded functions of \(t\). Using the following definition: A function of \({\mathbf s}= [s_j]_{n\times 1}\), \(s_j\in [t-\tau,t]\), \(Z({\mathbf s})\), is said to be in \(V({\mathbf x}(t))= {\mathbf x}^T(t) P{\mathbf x}(t)\), if there exists a scalar \(L>0\) such that \(V(Z({\mathbf s}))\leq LV({\mathbf x}(t))\), the paper proves the following Theorem: Suppose that \((A_1,B_1)\) is controllable, \((A_1,C_1)\) is observable, \({\mathbf x}({\mathbf s})\), \(\dot{\mathbf x}({\mathbf s})\), \(\ddot{\mathbf x}({\mathbf s})\) are in \(V({\mathbf x}(t))\), and \(f(\cdot)\) satisfies \[ \begin{aligned} f_j(\cdot) &\leq [g_j^{(1)}(t)+ g_j^{(2)}(t-\tau_1)+ g_j^{(3)}(t-\tau_2)+ g_j^{(4)}(t-\tau_3)] N_j/4,\\ g^{(r)}(s) &= [g_1^{(r)}(s), g_2^{(r)}(s),\dots, g_n^{(r)} (s)]^T, \quad r=1,2,3,4;\;j=1,2,\dots, n, \end{aligned}\tag{2} \] and \(g^{(r)}(s)\) is in \(V({\mathbf x}(t))\), where \(s\in [t-\tau,t]\), \(\tau= \max\{\tau_{ij}\mid r=1,2,3,4\); \(j=1,2,\dots, n\) (or \(m)\}\), \(N_j\geq 0\) are constant, \(N=\max (N_j)\). Then, there exist two positive numbers \(\Delta_1>0\) and \(\Delta_2>0\) such that when \(0\leq \tau< \Delta_1\), \(0\leq N< \Delta_2\), the zero solution to the following closed-loop system of system (1) using optimal control law \({\mathbf u}(t)= -K{\mathbf x}(t)\) is asymptotically stable: \[ \begin{aligned} \dot x(t)&= (A_1-B_1K){\mathbf x}(t)+ A_2[{\mathbf x}(t-\tau_1)-{\mathbf x}(t)]+ A_2[{\mathbf x}(t-\tau_2)-{\mathbf x}(t)]\\ &\quad +B_2K[{\mathbf u}(t-\tau_3)-{\mathbf u}(t)]+ f(\cdot). \end{aligned}\tag{3} \] {}

Recommendations

Cited in

(8)

This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4489657)