Cascaded iterative learning control for improved task execution of optimal control

摘要

Iterative learning control (ILC) is used as a means for task execution of a time sub-optimal trajectory. The use of cascaded ILC procedures in a robot application is experimentally shown to drastically enlarge the region of convergence and efficiently compensate for unmodeled dynamics in the motion system. The paper deals with a control problem common in machines for packaging fluids. Objectives considered are the duration of motion, maximum liquid slosh during the motion, and the residual slosh after the motion.