3D Successive Nanoscale Interactions-Driven Mechanoreceptors with Broad Linear Range and Ultra-High Sensitivity for Efficient Gesture Recognition

Wearable biosensing systems for monitoring daily human activities have been extensively investigated by researchers of human–machine interactions. However, the transmission of biomechanical signals is degraded owing to limitations arising from the non-ideal characteristics of hardware, which obstructs the precise recognition of human motions by sensing systems and practical applications. In this study, we demonstrates, for the first time, 3D successive nanoscale interaction-driven artificial slow-adapting (SA) mechanoreceptors to develop a biosensing system integrated with sensing, learning, and computational functionalities for efficient sign-language detection and translation. Specifically, nanocellular graphene (GN) with a high conductivity of 1.26 × 10⁴ S m⁻¹ is constructed via submicron porous template-assisted synthesis. More importantly, by leveraging unique conductive networks driven by 3D successive nanoscale interactions of artificial SA mechanoreceptors, co-optimization of the sensitivity (gauge factor = 242) and linear operating range is achieved, unlike the case of foam GN sensors. An artificial biosensing system (ABSS) based on sensing arrays operating via machine-learning algorithms achieves real-time gesture detection and recognition with 98.8% recognition accuracy. This 3D successive nanoscale template-assisted design strategy may provide a new path toward high-performance ABSS fabrication, thus enabling potential applications in fields such as human–computer interaction, virtual reality, and healthcare.