Machine Learning‐Enhanced Smart Interactive Glove Utilizing Flexible Gradient Ridge Architecture Iontronic Capacitive Sensor

Abstract

ABSTRACT The rapid advancement of artificial intelligence has spurred significant advances in wearable systems based on flexible sensors, which nevertheless still face significant challenges in maintaining high sensitivity and stability over broad pressure ranges. Here, an iontronic capacitive pressure sensor featuring a gradient ridge architecture (GRA) electrode integrated with a polyvinyl alcohol‐phosphoric acid (PVA‐H 3 PO 4 ) ionic gel film is developed. The synergistic combination of the engineered electrode morphology and the mechanical durability and electrochemical stability of the ionic gel film enables a high sensitivity of 5.12 kPa −1 in the low‐pressure region (0–20 kPa) and maintains a linear sensitivity of 2.02 kPa −1 over 20–463 kPa. The sensor further demonstrates rapid response and recovery times of 36.5 and 38.6 ms, respectively, along with excellent durability over 10 000 loading cycles. When integrated into an intelligent interactive glove, the GRA sensor enables adaptive robotic hand control and improves reliability through a tactile feedback strategy. Leveraging a 2D convolutional neural network, the smart assistive communication system achieves 97% recognition accuracy for ten distinct hand gestures, enabling real‐time translation of intuitive gestures into personalized text and voice outputs. These findings highlight significant potential for next‐generation human–machine interfaces, intelligent robotics, and high‐performance wearable electronics.