Application study on Iterative Learning Control of high speed motions for parallel robotic manipulator

摘要

This paper presents a novel application of Iterative Learning Control (ILC). It is about bettering control performance of Parallel Kinematic Manipulators (PKM) in the range of high dynamics. Such mechanisms suffer very often from lack of accuracy at high speed, since uncertainties, nonlinearities and disturbances have an important impact. The case seems to be predestinated for applying ILC. This will be demonstrated in this paper, where additional to a feedforward decoupling control structure, ILC techniques are used to decrease remaining tracking errors. Three algorithms are chosen to be validated, adjusted and compared. It is shown, that with an appropriate strategy, linear ILC approaches can be implemented on highly nonlinear and coupled MIMO-Systems, such as parallel manipulators.