Integrated Molecular‐Structural Design of “Mille‐Feuille” Triboelectric Sensors for Self‐Powered Tactile Perception and Material Recognition

ABSTRACT Achieving human‐skin‐like multidimensional tactile perception in self‐powered electronics remains fundamentally constrained by the intrinsic trade‐offs among sensitivity, mechanical compliance, and electrical output. To overcome these bottlenecks, a biomimetic “mille‐feuille” triboelectric sensor is engineered via hierarchical nanofibrous membranes, transitioning conventional 2D interfacial contact into 3D volumetric nesting to significantly increase the effective contact area. Incorporating a zwitterionic polymer poly(sulfobetaine methacrylate) (PSBMA) into the triboelectric matrix enables multiscale enhancement of dielectric polarization, charge trapping, interfacial contact, and mechanical robustness. This integrated design concurrently improves surface charge density and contact area, culminating in a remarkable 2.16‐fold enhancement in triboelectric performance compared to pristine counterparts. The sensor exhibits an ultra‐low detection limit (1.21 Pa), a broad working range (2.45–100 kPa), fast response and recovery times (35 and 29 ms), and excellent durability over 10 000 cycles. Beyond conventional motion monitoring, integrating this sensor with a robotic gripper and a convolutional neural network (CNN) enables active material cognition. The system demonstrated accurate recognition and sorting of geometrically identical objects, reaching 100% test accuracy under current dataset and sorting of geometrically identical objects based on their material‐dependent tactile responses during physical interaction. This work provides a robust material‐algorithm synergistic paradigm for intelligent robotics and tactile sensing applications.