A Graphene Oxide-Based Tribo-Piezoelectric Hybrid Sensor for Armed Swimming Action Recognition

Abstract

Limb movement analysis provides a convenient and effective approach for physiological safety monitoring and motion standardization assessment of personnel working in humid, wading, or underwater environments. However, in wet and wading environments, the existing wearable flexible sensors generally suffer from several limitations such as insufficient sensitivity, rapid performance degradation, and poor environmental stability. In this paper, we propose a corrosion-resistant tribo-piezoelectric hybrid sensor (CRTPHS) primarily composed of an intrinsically corrosion-resistant fluorinated graphene oxide film with a micro-truncated pyramid array structure (MFGOF) and gold-coated polydimethylsiloxane-barium zirconate titanate (PDMS-BZT). The CRTPHS exhibited a sensitivity of up to 1.86 V/N, a response time of 106 ms, a peak power density of 1.98 mW/m2, and stable output performance after 50,000 cycles. In corrosion resistance tests, the CRTPHS retained a sensitivity of 1.27 V/N and excellent underwater sensing capability even after 3 days of seawater immersion. By integrating a lightweight random forest (RF) algorithm, the CRTPHS enabled recognition of five simulated armed-swimming motion states, achieving an average cross-subject accuracy of 99.25% ± 0.07% under leave-one-subject-out cross-validation (LOSO-CV). This study not only provides a novel fabrication method for developing corrosion-resistant self-powered sensors but also offers a new strategy for intelligent posture recognition in corrosive seawater environments.