Optimality conditions for implicit controlled systems

From MaRDI portal





We give necessary and sufficient conditions for optimality in the optimal control problem for a nonlinear implicit controlled system. An implicit controlled system is described by the following set of equalities and inequalities that must be satisfied by the trajectory \(x(t)\) corresponding to the chosen control \(u(t)\): \[ \begin{alignedat}{4} \varphi_i(\dot x(t),x(t),u(t)) &\leq 0, &\quad i&\in I_\varphi^<, &\qquad \varphi_i(\dot x(t),x(t),u(t)) &= 0, &\quad i&\in I_\varphi^=,\\ g_i(x(t),u(t)) &\leq 0, &\quad i&\in I_g^<, &\qquad g_i(x(t),u(t)) &= 0, &\quad i&\in I_g^=, \end{alignedat} \] where \(\varphi_i(v,x,u)\), \(i\in I_\varphi\), \(g_i(x,u)\), \(i\in I_g\), are continuous real-valued functions, \(I_\varphi= I_\varphi^<\cup I_\varphi^=\), \(I_g= I_g^<\cup I_g^=\), \(I_\varphi^<\) and \(I_\varphi^=\) as well as \(I_g^<\) and \(I_g^=\) are finite pairwise disjoint sets of indices. The approach taken here is based on considering the optimal control problem for the implicit controlled system as an extremal problem for the differential inclusion \[ \dot x\in F(x),\qquad x(0)=x_0. \] The multi-valued mapping \(F(x)\) can be represented as \[ F(x)= \bigcup_{u\in U(x)} F(x,u), \] where the multi-valued mappings \(F(x,u)= \{v: \varphi(v,x,u)\leq 0\}\), \(U(x)= \{u\in U: g(x,u)\leq 0\}\) are defined in terms of the nonsmooth functions \[ \begin{aligned} \varphi(v,x,u) &=\max\{0, \max_{i\in I_\varphi^<} \varphi_i(v,x,u), \max_{i\in I_\varphi^=}|\varphi_i(v,x,u)|\},\\ g(x,u) &=\max\{0, \max_{i\in I_g^<} g_i(x,u), \max_{i\in I_g^=}|g_i(x,u)|\}. \end{aligned} \] .











This page was built for publication: Optimality conditions for implicit controlled systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q676808)