Non-uniform spiking touch sensor for classification of movement patterns using neural networks

Non-uniform spiking touch sensor for classification of movement patterns using neural networks

Expert Systems with Applications

2026

https://ruj.uj.edu.pl/handle/item/556244 We propose a new capacitive touch sensor design with a dielectric with non-uniform electrical permittivity. A sensor of that kind, connected to a signal conditioning circuit, generates a spiking signal that encodes spatial information. The sensor behaves similarly to a slowly-adapting type I (SA-I) first-order tactile neuron that provides tactile information to a human neural system. We achieved over 97% accuracy in recognising the movement patterns.

128866

295

https://doi.org/10.1016/j.eswa.2025.128866

https://www.sciencedirect.com/science/article/pii/S0957417425024832

ZAMKNIĘTY