Discretization methods for optimal control problems with state constraints (Q2490715)

Control systems described by nonlinear ordinary differential equations are considered. A cost functional and constraints on the state and on the control are defined. The state constraints include pointwise state constraints. The existence of a solution is shown for different formulations of the problem. Necessary conditions similiar to the classical minimum principle are derived. The continuous problem is then discretized. For a given discrete control the state is calculated by the implicite midpoint rule assuming a linear control on each union of two successive intervals. Necessary conditions for optimality are derived for the discrete problem. The behaviour of the discrete solution for refined discretization is then investigated. It is shown that the discrete solution is converging in the limit to the continuous solution. An algorithm for solving the optimization problem based on gradient methods is presented. The proposed method is illustrated by numerical examples.