Soft Robotic Glove with Sensing and Force Feedback for Rehabilitation in Virtual Reality

Many diseases, such as stroke, arthritis, and spinal cord injury, can cause severe handimpairment. Treatment options for these patients are limited by expensive hand rehabilitation devicesand dull treatment procedures. In this study, we present an inexpensive soft robotic glove for handrehabilitation in virtual reality (VR). Fifteen inertial measurement units are placed on the glove forfinger motion tracking, and a motor-tendon actuation system is mounted onto the arm and exertsforces on fingertips via finger-anchoring points, providing force feedback to fingers so that the userscan feel the force of a virtual object. A static threshold correction and complementary filter are usedto calculate the finger attitude angles, hence computing the postures of five fingers simultaneously.Both static and dynamic tests are performed to validate the accuracy of the finger-motion-trackingalgorithm. A field-oriented-control-based angular closed-loop torque control algorithm is adopted tocontrol the force applied to the fingers. It is found that each motor can provide a maximum forceof 3.14 N within the tested current limit. Finally, we present an application of the haptic glove ina Unity-based VR interface to provide the operator with haptic feedback while squeezing a softvirtual ball.