Learning control schemes for reference tracking of under-actuated systems

Learning control systems can significantly improve the tracking performance and robustness of repetitive systems. Specifically, the iterative learning control (ILC), which follows the concept of learn from mistakes, can offer precise trajectory tracking through adaptive framework. This talk will explain the current cycle feedback ILC for a class of under-actuated system and analyze the potentials of augmenting ILC with feedback controller through hardware in loop (HIL) testing. Moreover, a brief overview on the design of reinforcement learning control for reference tracking and vibration suppression of rotary flexible link will be presented.