\({\mathcal H}_\infty\) constant gain state feedback stabilization of stochastic hybrid systems with Wiener process (Q2387476)

The authors consider the stabilization problem for the following class of stochastic differential and algebraic equations \[ dx(t)= [A(r_t)x(t)+ B(r_t)u(t)+ B_\eta(r_t) \eta(t)]dt+ \sigma(r_t)d\xi(t),\;x(t_0)= x_0,\;t\in [0,+\infty),\tag{1} \] \[ y(t)= C_y(r_t) x(t)+ D_y(r_t) u(t)+ B_y(r_t) \eta(t),\tag{2} \] \[ z(t)= C_z(r_t) x(t)+ D_z(r_t) u(t)+ B_z(r_t) \eta(t),\tag{3} \] where \(x\in\mathbb{R}^n\) is the state vector, \(x_0\in\mathbb{R}^n\) is the initial state, \(y\in\mathbb{R}^{n_y}\) is the measured output, \(z\in\mathbb{R}^{n_z}\) is the controlled output, \(u\in\mathbb{R}^m\) is the control vector, \(\eta\in\mathbb{R}^l\) is the system external disturbance, \(\xi(t)\) is a standard Wiener process independent of the continuous-time Markov process \(\{r_t,t\geq 0\}\) taking values in the finite space \(S= \{1,\dots, N\}\) that describes the evolution of the mode at time \(t\) and when \(r_y= i\), \(A(i)\), \(B(i)\), \(B_\eta(i)\), \(\sigma(i)\), \(C_y(i)\), \(D_y(i)\), \(C_z(i)\), \(D_z(i)\) and \(B_z(i)\) are matrices with appropriate dimensions. The authors study the \(H_\infty\) stabilization problem by designing a state-feedback controller with constant gain that stabilizes in stochastic sense the system (1)--(3). This controller rejects also disturbances with a derired level \(\gamma> 0\). Using Lyapunov technique, LMI-based conditions are derived. Two numerical examples illustrate the obtained results.