Dermis‐Inspired Flexible Capacitive Sensor with High Sensitivity and Wide Linear Range for Material Identification

Abstract Traditional flexible capacitive sensors struggle to balance high sensitivity with a wide linear range due to the lack of a biomimetic hierarchical structure in their single‐scale microstructures. Inspired by the gradient microstructures of human skin, particularly the microfolds of the basement membrane and the interlaced papillae of the dermis‐epidermis junction, this paper proposes a multi‐scale microstructure capacitive sensor that combines ultraviolet picosecond laser‐induced graphene electrodes with an ordered‐disordered dielectric layer design. The sensor achieves high sensitivity and wide linear range (0–500 kPa) across the entire pressure range through a dual mechanism: sandpaper‐replicated random fractal roughness induces multi‐point stress concentration, simulating the amplification effect of skin micro‐wrinkles; and dual‐height micro‐protrusions with a stepped compression structure enable capacitive relay control. The sensor features ultra‐fast response/recovery capability (50 ms), 27 500‐cycle stability, and a microforce detection limit of 0.81 g. Combined with a multilayer perceptron (MLP)‐based machine learning model, the platform achieves 98.8% accuracy in hardness recognition across six materials spanning from sponge to titanium alloy, advancing robotic tactile perception toward human‐level cognition.