Assisting control for pendulum-like juggling in human adaptive mechatronics

Abstract

Human Adaptive Mechatronics consists of human and mechatronics, and aims to develop mechatronics systems adapting and assisting human operations. In this paper, an operation assisting system is presented, and the effectiveness was confirmed experimentally using a haptic interface device for the pendulum-like juggling operation, which is appropriate for evaluation of an assistance control method since it requires human voluntary motion control inevitably used in a human–machine interface operation. The presented assistant control system consists of three control sub-systems: a virtual-internal-model force controller, a servo controller to match the response of the controlled machine to the virtual model, and the correction force controller to assist the human operator. Applying the presented control to the juggling task experimentally, it was demonstrated that not only assistant to the operator but also recovering against human inadequate operation could be realized without switching the system structure. Additional identification analysis of the trained operator’s frequency characteristic proved that the operator had mastered the interface-device manipulation because the cross-over model property which is a sign of a skilled operator was confirmed.