Abstract
Robot-assisted rehabilitation has been developed and proved effective for motion function recovery. Humanization is one of the crucial issues in the designing of robot-based rehabilitation system. However, most of the previous investigations focus on the simplex position control when comes to the control system design of robot-assisted passive training, and pay little attention to the dynamic adjustment according to the patient’s performances. This paper presents a novel method to design the passive training system using a developed assessing-and-regulating section to online assess the subject’s performances. The motion regulating mechanism is designed to dynamically adjust the training range and motion speed according to the actual performances, which is helpful to improve the humanization of the rehabilitation training. Moreover, position-based impedance control is adopted to achieve compliant trajectory tracking movement. Experimental results demonstrate that the proposed method presents good performances not only in motion control but also in humanization.