Electron Compensation Enhanced Triboelectric Sensor Assisted by Machine Learning for Tactile Perception Recognition

作者：Xiangkun Bo, Wei Wang, Hong Zhao, Ruiqin Zhang, Walid A. Daoud · 发表于：Advanced Functional Materials · 年份：2025 · DOI：10.1002/adfm.202514567 · 被引用次数：4 · 研究领域：Advanced Sensor and Energy Harvesting Materials、Tactile and Sensory Interactions、Conducting polymers and applications

Abstract Artificial intelligence‐assisted self‐powered triboelectric sensors have significant potential in sensing and bionic engineering, however their low electrical output limits their sensitivity and signal‐to‐noise ratio. Herein, a novel electron compensation strategy is proposed to enhance the electrical output. The approach involves integrating polyethyleneimine (PEI) and carbon black in a polymeric matrix to introduce an abundance of electron‐donating groups and thus establishing an electron‐transportation network. This facilitates the extraction of bulk electrons to replenish the surface electron consumption during contact‐separation. The carbon black serves to buffer the tearing force during stretching, thereby enhancing the mechanical durability of the film. Furthermore, the incorporation of carbon black creates electron conductive networks, enabling efficient transport of electrons from the bulk of PEI to the surface. As a result of high tribopositivity and electron replenishing mechanism, the composite‐based self‐powered triboelectric sensor shows notable open circuit voltage of 258 V and charge density of 0.270 mC m −2 , which are 3.85‐fold and 4.09‐fold higher than pristine water polyurethane, demonstrating a high sensitivity of 75.3 V mm −1 . This research provides a novel approach toward high sensitivity for the advancement of machine learning assisted tactile perception recognition.