Controllability under positive constraints for quasilinear parabolic PDEs (Q2119441): Difference between revisions

A class of quasilinear parabolic equation is considered in a bounded spatial domain~\(\Omega\subset\mathbb R^N\) and in a temporal interval~\((0,T)\), under homogeneous Dirichlet boundary conditions. The control is internal. The control input~\(v\) is nonnegative~\(v(x,t)\ge0\) and essentially bounded~\(v\in L^\infty(\Omega\times(0,T))\); the control operator is the indicator function of a given nonempty open subset~\(\omega\subset\Omega\). Given a trajectory~\(\overline y(x,t)\) with initial state~\(\overline y(0)=\overline y_0\) and corresponding control~\(\overline v(x,t)\ge\eta>0\), the goal is to steer the state~\(y\) of the system, issued from a different initial state~\(y(0)=y_0\ne\overline y_0\), to~\(\overline y(T)\) at time~\(t=T\) by using an appropriate nongegative control~\(v(x,t)\ge0\). Under some regularity assumptions, it is shown that the goal can be reached for arbitrary given~\(y_0\ne\overline y_0\) and for large enough~\(T\ge T_{\min}(y_0)\). It is also shown that necessarily \(T_{\min}(y_0)>0\) (cf. Thm.~2.4).