Stability of solutions to an optimal control problem for a continuous Fornasini-Marchesini system (Q2702525)

The paper considers a sequence of optimal control problems for the equations NEWLINE\[NEWLINEw_{xy}= g_k(x,y,w, w_x, w_y,u),\quad k= 0,1,2,\dots,NEWLINE\]NEWLINE in the unit square \(K= (0,1)\times (0,1)\) with controls NEWLINE\[NEWLINEu\in U_k= \{u\in L_2(K)\mid u(x,y)\in M_k\subset R^m\},\quad k= 0,1,2,\dots\;.NEWLINE\]NEWLINE It is shown that if the functions \(g_k\) (together with the functions which give the boundary conditions and the integrand of the cost functional) converge in an appropriate sense to the limit function \(g_0\) and the sets of \(M_k\) converge in the Hausdorff metric to the set \(M_0\), then the sets of solutions of the corresponding sequence of optimal control problems converge to the set of solutions of the limit problem.NEWLINENEWLINEFor the entire collection see [Zbl 0952.00062].