Reliable linear-quadratic control for singularly perturbed systems

From MaRDI portal





This paper studies the reliable linear-quadratic (LQ) state-feedback control for singularly perturbed systems. The problem under consideration is described by: \[ \begin{aligned} \dot x_1 & = A_{11}x_1 + A_{12}x_2+B_1u,\quad x_1(0) =x_{10}\\ \varepsilon \dot x_2 & = A_{21}x_1 + A_{22}x_2 +B_2u,\quad x_2(0)=x_{20}\end{aligned} \] where \(x_i\in\mathbb{R}^{n_1}\), \(u\in \mathbb{R}^m\), \(A_{22}\) is invertible, \(\varepsilon > 0\) is a small parameter and the cost function is given by \[ J=\int_0^\infty (x^T_1 Q_1x_1+ x^T_2 Q_2 x_2+u^T Ru)dt \] \(Q_1\geq 0\), \(Q_2\geq 0\), \(R > 0\). By time-scale decomposition, the full-order system is decomposed into the slow and fast subsystems. The reliable composite control law is given based on the designs of the two subsystems. The designed composite control guarantees the robust stability and LQ performance of the full-order closed-loop system despite any actuator outages within a selected subset of actuators. An illustrative example is presented to demonstrate the approach addressed.











This page was built for publication: Reliable linear-quadratic control for singularly perturbed systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4796856)