A method for unified optimization of systems and controllers

A unified method for solving control system optimization problems is suggested. All system matrices are allowed to be functions of the design variables. The method makes use of an implementation of a sequential quadratic programming algorithm (NLPQL) for solution of general constrained nonlinear programming problems. It is shown how to compute the gradients of the objective function and the constraint functions imposing eigenvalue constraints. In an example it is demonstrated how the method can solve a high-dimensional problem, where the initial condition covariance assumption is used to ensure constant initial mechanical energy in the beam during all iterations in the optimization