Rapid exact controllability of the wave equation by controls distributed on a time-variant subdomain (Q1284268)

This paper deals with the rapid exact controllability of the following wave equation with a locally distributed control: \[ \begin{cases} y_{tt}-\Delta y=\chi_{G(t)} (x)u(x,t)\quad & \text{in } \Omega\times(0,\infty), \\ y=0\quad & \text{on }\partial \Omega\times [0,\infty),\\ y(x,0)=y_0(x),\;y_t(x,0)=y_1(x) \quad & \text{in }\Omega,\end{cases} \tag{1} \] where \(\Omega\subset \mathbb{R}^n\) is a bounded domain with the Lipschitz boundary \(\partial\Omega\), \(G(t)\) is a subdomain of \(\Omega\) for each \(t\in[0,\infty)\) and \(\chi_{G(t)}(x) u(x,t)\) is the control. Let be \(U=L^2_{\text{loc}} (0,\infty; L^1(\Omega))\) and \({\mathcal G}\) a family of set-valued functions \(G(\cdot)\) defined on \([0,\infty)\) taking subdomains of \(\Omega\) as its values. Then the system (1) is said to be rapidly exactly controllable if for any \(T>0\), there exists a \(G(\cdot)\in{\mathcal G}\) such that for any \((y_0,y_1)\in H^1_0(\Omega)\times L^2(\Omega)\), there exists a control \(u(\cdot)\in U\) such that the corresponding solution \(y(x,t)\) of (1) satisfies \[ y(x,T)=y_t(x,T)=0,\quad\text{a.e. }x\in\Omega. \] The main result states that if any \(G(\cdot)\in{\mathcal G}\) is continuous with respect to the Lebesgue measure and for any \(T>0\), there exists a \(G(\cdot)\in{\mathcal G}\) and \(\sigma\in(0,T)\) such that \[ D_\sigma\subseteq \bigl\{(x,t)\in \Omega_T\mid x\in G(t),t\in[0,T]\bigr\}, \] then (1) is rapidly exactly controllable under \({\mathcal G}\), where \[ D_\sigma=(K_\sigma \times\mathbb{R}^{n-1})\cap\Omega_T, \quad\Omega_T= \Omega\times (0 ,T), \] \[ K_\sigma= \bigl\{(x_1,t)\in [0,\beta]\times [0,T]\mid ax_1<t<ax_1+ \sigma\bigr\}, \] \[ \begin{aligned} a= & {T-\sigma \over\beta},\\ \beta = & \sup\bigl\{x_1 \in\mathbb{R} \mid\exists x'\in \mathbb{R}^{n-1},\quad \text{such that }(x_1,x') \in\Omega\bigr\}. \end{aligned} \] The problem of the rapid exact controllability is reduced to checking a certain estimate and the proof of the mentioned result is based on the uniqueness-compactness argument.