Iterative learning control for robot manipulators

Abstract

When a system is performing the same task repeatedly it is, from an engineering perspective, advantageous to use the knowledge from the previous iterations of the same task in order to reduce the error on successive trials. In control systems, the aim is to force the system output to follow a desired trajectory as closely as possible. Specific norms and measures of optimality are used to determine how close the output is to the desired trajectory. Although control theory provides many different possible solutions for such problem, it is not always possible to achieve a desired set of performance requirements. This may be due to the presence of unmodeled dynamics or parametric uncertainties exhibited during the system operation, or due to the lack of suitable design techniques for particular class of systems. Iterative learning control (ILC) is a relatively new addition to these techniques that, for a particular class of problems, can be used to overcome some of the difficulties associated with performance design of control systems.