Upper semicontinuity of the solution map to a parametric boundary optimal control problem with unbounded constraint sets (Q2673532)

Given \(\Omega \subset\mathbb{R}^2\) a bounded set whose boundary \(\Gamma\) is \(C^{1,1}\) the authors consider the following problem: find \(u \in L^2(\Gamma)\) and \(y \in H^1 (\Omega) \cap C(\overline\Omega)\) such that a functional \[ J (y, u, \mu) \text{ is minimum} \] subject to \[ \begin{cases} A y + f(x,y) = 0 & \text{ in } \Omega \\ \partial_{n_A} y = u + \lambda^{(1)} & \text{ in } \Gamma \end{cases} \] \[ g(x,y) + u (x) + \Lambda^{(2)} \leqslant 0 \qquad \text{ in } \Gamma. \] Here \[ J (y, u, \mu) = \int_{\Omega} L(x,y(x), \mu^{(1)} (x)) \, dx + \int_{\Gamma} \big(\ell (x, y (x), \mu^{(2)}) + \varphi (\mu^{(2)}) u^2 (x) \big) \, dx \] with \(L : \Omega \times \mathbb{R} \times\mathbb{R} \to\mathbb{R}\), \(\ell : \Omega \times\mathbb{R} \times \mathbb{R} \to\mathbb{R}\), \(\varphi:\mathbb{R} \to \mathbb{R}\), \(f : \Omega \times \mathbb{R} \to \mathbb{R}\), \(g : \Gamma \times \mathbb{R} \to \mathbb{R}\), and \((\mu, \lambda) \in \big(L^{\infty} (\Omega) \times L^{\infty} (\Gamma) \big) \times \big(L^{\infty} (\Gamma) \big)^2\), \begin{align*} A y &= - \sum_{i, j = 1}^2 D_j \big(a_{ij} D_i y \big) + a_0 y, \\ \partial_{n_A} y &= \sum_{i, j = 1}^2 a_{ij} D_i y \, \nu_j, \quad \nu \text{ outward normal to } \Gamma. \end{align*} Once denoted by \(P (\mu, \lambda)\) this problem, \((\bar\mu, \bar\lambda)\) the reference point, \(P (\bar\mu, \bar\lambda)\) the unperturbed problem and \(S (\mu, \lambda)\) the solution set of problem \(P\), the authors prove that \(S\) is upper semicontinuous and, if \(S (\bar\mu, \bar\lambda)\) is a singleton, \(S\) is continuous at \((\bar\mu, \bar\lambda)\).