Highly Stretchable Serpentine-Shaped Strain Sensor Based on 3-D Printing for Gestures Recognition

Abstract

The gesture recognition technology based on flexible strain sensors has attracted widespread research interest in fields such as human–computer interaction. However, the complex fabrication process of current flexible strain sensors limits their mass production capacity. Herein, we introduce a fabrication method for strain sensors based on 3-D printing technology. A custom serpentine-shaped flexible strain substrate was constructed using low-cost thermoplastic polyurethane (TPU) elastomer as printing material, which met the needs of mass production of the flexible substrate and improvement of mechanical properties of the sensor. Soaking the substrate in a specific proportion of N, N-dimethylformamide (DMF)/carbon black (CB) solution, the sensing layer based on a stable conductive network was constructed on its surface using ultrasound technology, and further softening enhanced its deformation ability. The proposed strain sensor exhibits excellent sensing performance with a wide strain range of up to 200%, high sensitivity of 58.08, fast response time of about 0.1 s, a nd high durability (1000 cycles under 50% strain), achieving detection of human joint motion status. Finally, a wearable gesture recognition system was established based on the data glove integrated with flexible sensors and a support vector machine (SVM), achieving an accuracy rate of 96% for recognizing ten commonly used gestures and translating them into audible speech in real-time. The experimental results demonstrate the practical value and potential application of the designed sensor in wearable devices, and to build communication channels for people with language disorders in the future.