Iterative learning identification

An iterative learning identification method is proposed for curve identification problems. The basic idea is to convert the curve identification problem into an optimal tracking control problem. The measured trajectories are regarded as the desired trajectories to be optimally tracked and the curve to be identified is taken as a virtual control function. A high-order learning updating law is applied. A convergence condition is obtained in a general problem setting. Two case studies, which are related to the aerodynamic drag coefficient curve extraction from actual flight testing data, are presented to show the practical usefulness of the proposed method.