A remark on dynamic programming with final state constraints (Q796093)

A controlled system is described by a set of nonlinear ordinary differential equations in \(R^ n\). The control parameter is taking values in a finite or in a compact set. The penalty function is in the usual integral form with discounting factor plus penalty at the time of exit from a given domain. The solution of the Hamilton-Jacobi equation is shown to coincide with the limit of solutions with altered boundary conditions.