An Iterative Learning Control with a Reduced Input Update Rate

In this paper, a simple and effective multirate iterative learning control (ILC) scheme, pseudo-downsampled ILC, is proposed to obtain good learning transient and deal with initial state error. In the proposed scheme, error and input signals collected from feedback control system are downsampled before they are used in ILC learning law. The output of ILC is interpolated to generate fast rate input for the next cycle. By downsampling to make certain condition holds, exponential decay of tracking error can be guaranteed. Other advantages include that it needs not a filter design; it reduces memory size and computation burden; it generate better tracking accuracy than conventional ILC. The design method is provided and experimental results are presented to verify the proposed scheme.