Wearable Human–Machine Gesture Interaction Based on Fabric Piezoelectric Sensor

Gesture recognition has great application prospects in the field of human-machine interaction (HMI). To date, the development of wearable and cost-effective sensors for gesture recognition remains a challenge. In this study, we propose a comprehensive system solution for gesture recognition and HMI-based on electrospun polyvinylidene fluoride (PVDF)/barium titanate (BaTiO3) nanocomposite membrane sensors. The sensor comprising three types of fabric can effectively recognize different bending angles and exhibits reliable repeatability. The gesture classification recognition based on the sensor exhibits high accuracy in three machine learning algorithms [K nearest neighbor (KNN), support vector machine (SVM), and neural network (NN)], with accuracy rates of 96.7%, 97.9%, and 97.8% achieved over 5000 groups of ten different gestures. Additionally, we design a scheme for real-time remote control of a mechanical claw based on this sensor. The results demonstrate that the proposed solution provides a promising platform for gesture recognition and HMI.