A Compact Gesture Sensing Glove for Digital Twin of Hand Motion and Robot Teleoperation

Capture of hand gestures using data gloves has a wide range of applications in human–machine interaction, e.g. virtual-reality manipulation and robot teleoperation. In this article, a compact wearable sensing glove is developed using largely deformed flexible beams embedded with strain gauges. An analytical model that exactly characterizes the force-deflection behavior is established to map the local deflections of the flexible beam to the geometric constraint induced by the finger tip. In our design, ordinary carbon-fiber sheets and industry-grade strain gauges are used to make the sensing glove compact in structure and reliable for measurement. A neural network is trained using model-based simulations to run the sensing algorithm in real-time, from the strain gauges’ outputs. A prototype is developed using easy-to-access materials and fabrication methods for experimental validation. The results show that the proposed sensing glove is able to capture finger motion through model-based sensing of accurate joint angle, with high compactness and sensor reliability. Demonstrations are provided to exhibit the potential of the developed sensing glove in the digital twin of hand motion and teleoperation of dexterous robotic hands.