The goal of this work is to develop a control framework to provide assistance to the subjects in such a manner that the interaction between the subjects and a robot-assisted rehabilitation system is smooth during the rehabilitation therapy. In order to achieve smoothness of interaction, a control framework is designed in such a way that it can automatically adjust the control gains of the robot-assisted rehabilitation system to modify the interaction dynamics between the system and the subject. An artificial neural network (ANN)-based proportional–integral (PI) gain scheduling controller is proposed to automatically determine the appropriate control gains for each individual subject. The human arm model is integrated with the ANN-based PI gain scheduling controller where the ANN uses estimated human arm parameters to select the appropriate PI gains for each subject such that the resultant interaction dynamics between the subject and the robot-assisted rehabilitation system could result in smooth interaction. Experimental results involving unimpaired subjects on a PUMA robot-based rehabilitation system are presented to demonstrate the efficacy of the proposed ANN-based PI gain scheduling controller on unimpaired subjects.
Read full abstract