Tactile Neuromorphic System: Convergence of Triboelectric Polymer Sensor and Ferroelectric Polymer Synapse.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 37611149.
- Also identified by DOI 10.1021/acsnano.3c05337.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Sensory neuromorphic systems are a promising technology, because they can replicate the way the human peripheral nervous system processes signals from the five sensory organs. Despite this potential, there are limited studies on how to implement these systems on a hardware neural network platform. In our research, we propose a tactile neuromorphic system that uses a poly(dimethylsiloxane) (PDMS)-based triboelectric sensor and a molybdenum disulfide (MoS<sub>2</sub>)/poly(vinylidene fluoride-trifluoro ethylene) (P(VDF-TrFE)) heterostructure-based ferroelectric synapse. The triboelectric sensor mimics a human tactile organ by converting tactile stimuli into electrical signals in real time. The ferroelectric synapse we developed demonstrates exceptional long-term potentiation/depression characteristics with a maximum dynamic range of 78 and a symmetrical value of 4.7. To assess the practicality of our proposed system, we conducted training and recognition simulations using Morse code alphabets and MNIST handwritten digits. The maximum recognition rate that we achieved was 96.17%.