Design, Modeling, and Optimal Position Control of a New Wrist Rehabilitation Robot Using the Stewart Platform

In this paper, a new application of the Stewart platform is developed for rehabilitation utilities. Stewart robot has a good workspace and load capacity similar to other parallel robots. Here the Stewart model is modified in order to be employed as a wrist rehabilitation robot. Both the direct and inverse kinematics of the robot is extracted using the Jacobian matrix. Afterward, the related dynamic model of the proposed robot is developed using the virtual work method. In order to keep the wrist of the patient within the desired path of rehabilitation, a nonlinear position control is designed and implemented using Computed Torque Method. In order to move the patient's wrist along a safe path, an optimal path is extracted through which, the least amount of acceleration and consequently interaction force will be implemented on the patient's wrist and this path is controlled using Computed Torque Method. With the aid of conducting some analytic simulations in MATLAB the applicability of the proposed robot for wrist rehabilitation is demonstrated. The verification of the model is performed by comparing the results with previous articles while the efficiency of the implemented controller is proved by comparing the actual path with the desired one in presence of disturbance. It is shown that with the aid of the proposed robot and controller, the wrist rehabilitation process of a patient can be successfully accomplished.