Realtime Hand-Gesture Recognition Based on Novel Charge Variation Sensor and IMU

Wearable hand gesture devices are of growing interest to enhance human-machine interfaces as well as digital communication between humans. Although many solutions have been proposed using various sensing technologies, the development of a truly wearable, low power, reliable and unobtrusive device is still a challenge. This work proposes a new device based on a combination of a novel electrostatic sensing circuit fused with a six degree of freedom Inertial Measurement Unit (IMU), to achieve a wearable bracelet capable of active gesture detection for over 11 hours on one battery charge. Due to the proposed tiny machine learning model, gestures are detected in real time by an onboard ARM Cortex M33 microcontroller with an accuracy of up to 87%. The IMU gesture recognition is enhanced by the electrostatic sensing bracelet, increasing accuracy by more than 10% with respect to IMU only estimations.