作者: Marco C. Campi , Toshiharu Sugie , Fumitoshi Sakai
关键词:
摘要: This paper presents a novel approach for the identification of continuous-time systems directly from sampled I/O data based on trial iterations. The method achieves through iterative learning control (ILC) concepts in presence heavy measurement noise. robustness against noise is achieved 1) projection signals onto finite dimensional parameter space and 2) Kalman filter type reduction. In addition, an alternative simpler given with some analysis. effectiveness demonstrated numerical examples nonminimum phase plant.