A Highly Sensitive Flexible Hydrogel Gesture Recognition System Based on Deep Learning

Gesture recognition based on the Internet of Things and artificial intelligence holds significant importance in human–computer interaction. Different from array-type hydrogel sensors, this research has developed a one-piece flexible hydrogel sensor based on an ion-cross-linked three-dimensional network structure. Through the design of the hydrogel components and its network structure, the Young’s modulus spanning from soft tissues to hard materials has been achieved. The hydrogel exhibits excellent output stability, surface adhesion, and skin adaptability. The test results show that the gesture recognition sensor prepared from this hydrogel has a minimum detection limit of only 0.04% and has excellent cyclic stability. On this basis, combined with the long short-term memory network (LSTM) deep learning algorithm, a complete intelligent gesture recognition system is constructed. This system can precisely recognize complex dynamic gestures and provide real-time feedback while simplifying the hardware connections and data flows. It offers an efficient solution for the field of intelligent care.