Interaction Method of Rehabilitation Robot Based on VPE and VEP

Abstract

The development of rehabilitation robots has shown significant potential in improving patient recovery through intelligent interaction mechanisms. The passage introduces two telerehabilitation methods called Vision-based Pose Estimation and Vision-based Environmental Perception. The first method is to apply a non-invasive solution in order to treat upper limbs by wearing a hand exoskeleton with human robot interaction. With a set of master-slave setup, the RGB-D camera captures an image of what operators perform and delivers it to the wearable hand exoskeleton remotely by a bidirectional combination link, while an external sensor records the forces. It is convenient to estimate data and find a better solution. The second method is about an exoskeleton robot that is equipped with the ability of environmental perception. It is surrounded by visional sensors, a control system and critical technologies. Vision-based Pose Estimation (VPE) and Vision-based Environmental Perception (VEP) are testified to be promising in the prospective telerehabilitation robot. This paper aims to provide a new perspective on human-robot interaction in the field of neurorehabilitation.