AI-driven signal enhancement for energy-efficient, wearable piezoelectric tactile sensing

This study presents the development of a fully flexible, low-power piezoelectric tactile
sensor based on the piezotronic effect, designed for applications in AI, augmented
reality, and smart wearable technologies. The sensor integrates piezoelectric materials
with organic field-effect transistors and β-polyvinylidene fluoride (PVDF) nanorod arrays,
 Full Article

  
6/3/26, 3:58 PM AI-driven signal enhancement for energy-efficient, wearable piezoelectric tactile sensing: Mechanics of Advanced Materials and Stru…
https://www.tandfonline.com/doi/full/10.1080/15376494.2026.2654074?scroll=top&needAccess=true 1/7
which convert mechanical stress into electrical signals. The AlN thin film, positioned
between titanium or molybdenum electrodes, efficiently amplifies the generated
piezoelectric voltage, significantly enhancing sensitivity and responsiveness. The
combined use of PVDF nanorods and AlN films boosts energy conversion efficiency,
enabling high precision in pressure detection. Experimental results show superior
pressure sensitivity, low detection limits, and rapid response times due to the unique
structural configuration and active material selection. Additionally, the device eliminates
the need for external gate voltage by employing piezotronic transistors to directly
convert mechanical stimuli into electrical outputs. For intelligent signal processing, the
system incorporates deep learning architectures—Convolutional Neural Networks
(CNNs), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRUs) with Multi�Head Attention—providing adaptive and high-fidelity interpretation of tactile data. This
innovation supports advancements in flexible electronic skin (e-skin) and next�generation human–machine interfaces.