Layer-by-Layer Self-Assembled Honeycomb Structure Flexible Pressure Sensor Array for Gait Analysis and Motion Posture Recognition with the Assistance of the ResNet-50 Neural Network

Abstract

With the rapid emergence of flexible electronics, flexible pressure sensors are of importance in various fields. In this study, a dopamine-modified melamine sponge (MS) was used to prepare a honeycomb structure of carbon black (CB)/MXene-silicone rubber (SR)@MS flexible pressure sensor (CMSM) through layer-by-layer self-assembly technology. Using SR as a binder to construct the honeycomb structure not only improves the mechanical properties of the sensor but also provides more attachment sites for CB/MXene, enhancing the stability of the conductive network. The honeycomb structure CMSM flexible pressure sensor exhibits high sensitivity (7.44 kPa^–1), a wide detection range (0–240 kPa), short response/recovery times (150 ms/180 ms), and exhibits excellent stability. In addition, a flexible smart insole has been developed based on a 6-unit CMSM sensor array, achieving plantar pressure detection. By combination of the ResNet-50 neural network algorithm with plantar pressure data under different postures, the recognition of 16 types of human motion postures has been achieved, with an accuracy rate of 90.63%. This study proposes a flexible sponge pressure sensor with excellent mechanical performance and sensing capabilities, providing new ideas and references for the design of flexible wearable sensor devices.