An Artificial Sensory Neuron with Tactile Perceptual Learning

TLDR

Sensory neurons in skin bridge external physical reality with tactile perception, enabling the organization, identification, and interpretation of sensory information through perceptual learning. The study aims to develop a neuromorphic electronic skin that integrates artificial intelligence for use in robotics and prosthetics. The system comprises sensing, transmitting, and processing modules analogous to a biological neuron, with a resistive pressure sensor converting pressure into electrical signals that are transmitted via a synaptic transistor through ionic/electronic coupling with a soft ionic conductor. The artificial sensory neuron recognizes spatiotemporal touch patterns, reducing the recognition error rate from 44 % to 0.4 % when combined with machine learning.

Abstract

Sensory neurons within skin form an interface between the external physical reality and the inner tactile perception. This interface enables sensory information to be organized identified, and interpreted through perceptual learning-the process whereby the sensing abilities improve through experience. Here, an artificial sensory neuron that can integrate and differentiate the spatiotemporal features of touched patterns for recognition is shown. The system comprises sensing, transmitting, and processing components that are parallel to those found in a sensory neuron. A resistive pressure sensor converts pressure stimuli into electric signals, which are transmitted to a synaptic transistor through interfacial ionic/electronic coupling via a soft ionic conductor. Furthermore, the recognition error rate can be dramatically decreased from 44% to 0.4% by integrating with the machine learning method. This work represents a step toward the design and use of neuromorphic electronic skin with artificial intelligence for robotics and prosthetics.