CONVERGENCE OF PSEUDOSPECTRAL METHODS FOR CONSTRAINED NONLINEAR OPTIMAL CONTROL PROBLEMS

摘要

In recent years, a large number of practical, nonlinear optimal control problems have been solved by pseudospectral methods. In an effort to better understand this new approach to solving control problems, we present convergence results for the discrete pseudospectral approximations of optimal control problems with mixed state and control constraints. A set of sufficient conditions are identified under which the solution of the discretized optimal control problem converges to the continuous solution at a spectral rate.