The SILP-Relaxation Method in Optimal Control: General Boundary Conditions II

In the first part of this paper the measure-theoretical approach to classical control problems, based on ideas of YOUNG in variational calculus and developed by RUB!O for control problems, was slightly extended by choosing a semi-infinite approach instead of a finite one. This results in a lower bound for the global minimum and an approximation for the optimal solution. It was still an open question, whether RUB tO's Approximation Theorem holds in the semi-infinite case and for more general boundary conditions. The second part of the paer dealswiththe discüisidñ of the approximation properties and gives as an example the numerical treatment of a nice geometric extremal problem by FOCKE.