Performance Evaluation of State and Output Feedback-based Repetitive Controllers for Rotational Control Systems

This study examines the tracking performance of state and output feedback-based repetitive controllers (RCs) for rotational control systems. First, a state feedback RC is designed in the time domain using a state-space model, which requires full state information for implementation. The state feedback RC is modeled as a 2D hybrid system, where the control and learning actions can be designed independently using Linear Matrix Inequality (LMI) solutions. This approach is evaluated on a rotational control system expressed as a strictly proper linear system with a relative degree of 1. Second, for comparison, an output feedback RC, also known as a general RC, is designed for the same system. Numerical experiment results show that while the state feedback RC achieves lower peak and root-means-quare errors, it exhibits slower transient response compared to the output feedback RC.