Iterative learning state estimation for nonlinear repetitive process

This paper explores the question about iterative learning observer design about a kind of nonlinear plants have repetitive operating characteristics. Different from traditional methods, the proposed iterative learning state observer is conducted and updated along the iteration direction. Furthermore, the proposed method has data-driven nature and derives from nonlinear systems directly, where no any model information is required except for the input and output measurements. A simulation case was employed to prove the performance of the given observer.