| The concept of Iterative Learning Control (ILC) was originated from the efforts to control systems that are required to perform repetitive tasks in the industrial field. The basic idea of ILC is that the information obtained from a previous trial is used to improve the control signal for the next trial until the desired performance level is reached. The iterative learning control has been developed and applied to many different fields, especially the robotic field. However, its application toward the hydraulic systems is rather sparse and is limited to a few articles. This thesis investigates the robust ILC of an electrohydraulic positioning system with a faulty actuator piston seal. The goal is to develop an ILC scheme that is tolerant to a faulty condition such as internal leakage. Toward this goal, three different aspects of iterative learning control are presented and compared, including the basic ILC, the ILC with proportional error feedback, and the ILC with current cycle feedback. (Abstract shortened by UMI.)... |
