Edinburgh Napier University

  Hand-motion detection technology has seen a rise in popularity over the years, due to the advancement of virtual reality (VR) technology and robotics. The medical industry has also shown interest in the technology for its ability to help detect hand dexterity issues and track the rehabilitation process of stroke victims with reduced hand motor functionality. Existing studies focus on hardware and filter design with little implementation of real-time data and wireless communication. This study advances the state of the art by developing a precise, robust data glove that combines flex sensors with inertial measurement units (IMUs) and streams data over Bluetooth Low Energy (BLE). In a Unity 3D environment, a digital hand model is driven in real time by the glove, faithfully replicating the wearer’s movements. The prototype achieved an accuracy of around 97.85% with a latency of roughly 0.28 s, capturing even high-frequency motion events. Given its performance and versatility, the proposed architecture shows strong promise for a wide range of applications, including hand dexterity related applications.