ReFlex, a haptic wrist interface for motor learning and rehabilitation

Abstract

Haptic interfaces have played an important role in the investigation of human factors, the identification of passive joint properties and gaining a better understanding of how the central nervous system learns to compensate for external dynamic perturbations through neuromuscular adaptation. Such devices further promise novel therapeutic approaches in neurorehabilitation, by assisting the patient to train functional tasks in an as-needed manner. With the aim of gaining deeper insights into the role of haptic feedback in motor learning and neurorehabilitation, we have developed a novel robotic wrist interface, the ReFlex, using a hybrid actuation approach, allowing to display a wider range of output impedances as well as a safe and efficient means of blocking the output for isometric training and rendering stable resistive force fields. This wrist module is mounted on a highly adjustable platform based on the consideration of constraints imposed by human factors and biomechanics. The system gives access and space for the placement of surface electrodes for electromyography of the main wrist and finger flexor/extensor muscle groups. We present the design of the ReFlex and analyze its performances in displaying forces and blocking the output. A method to increase the maximum brake torque and apparent output stiffness is described and validated.