Abstract
This demonstration presents a novel hardware sys-tem for high-speed, high-data-efficiency tactile sensing and real-time classification. The system uses a 16×16 high-impedance tactile sensor array and an event-driven strategy with a dynamic binary scanning algorithm to classify the digits. This approach leverages 99% data sparsity to achieve a 12.8× reduction in scan counts and a 38.2× data compression rate and a 28.4× dynamic range increase, significantly lowering system energy consumption. The resulting sparse data streams are processed by a multi-layer Convolutional Spiking Neural Network (Conv-SNN) on FPGA, which requires only 65% of the computation and 15.6% of the weight storage relative to a CNN, while maintaining 92.11% accuracy for real-time handwritten digit recognition. This live demo showcases a complete pipeline from analog tactile sensing to neuromorphic classification, highlighting the efficiency of event-driven systems for next-generation robotic perception and human-computer interaction. © 2026 IEEE.
| Original language | English |
|---|---|
| Title of host publication | 2026 IEEE International Symposium on Circuits and Systems (ISCAS) |
| Publisher | IEEE |
| Pages | 1679 |
| Number of pages | 1 |
| ISBN (Electronic) | 9798331577698 |
| ISBN (Print) | 9798331577704 |
| DOIs | |
| Publication status | Published - 2026 |
| Event | 2026 IEEE International Symposium on Circuits and Systems (ISCAS 2026): Circuits and Systems for Intelligent Society - Shanghai, China Duration: 24 May 2026 → 27 May 2026 https://2026.ieee-iscas.org/ |
Publication series
| Name | Proceedings - IEEE International Symposium on Circuits and Systems |
|---|---|
| ISSN (Print) | 0271-4302 |
| ISSN (Electronic) | 2158-1525 |
Conference
| Conference | 2026 IEEE International Symposium on Circuits and Systems (ISCAS 2026) |
|---|---|
| Abbreviated title | IEEE ISCAS 2026 |
| Place | China |
| City | Shanghai |
| Period | 24/05/26 → 27/05/26 |
| Internet address |
Bibliographical note
Author(s) information for this publication is provided by the author(s) concerned.Research Keywords
- Electronic Skin
- Event-Driven Sensing
- Neuromorphic Front-end
- Spiking Neural Network
Fingerprint
Dive into the research topics of 'Live Demonstration: An Event-Driven E-Skin System with Dynamic Binary Scanning and real time SNN Classification'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver