Artificial Skin and Electrotactile Stimulation for Advanced Tactile Feedback in Myoelectric Prostheses

Abstract

Modern prosthetic arms and hands are sophisticated robotic devices that can
\nprovide some of the motor functions lost due to an amputation. However,
\nnone of the commonly used commercial systems restores somatosensory
\nfeedback to its user. In principle, the latter can be achieved by recording data
\nfrom prosthesis sensors and conveying this information by stimulating the
\nsensory structures of the amputee using invasive and noninvasive interfaces.
\nMany such systems have been presented in the literature with promising
\nresults; however, they all rely on using several stimulation points to transmit
\ninformation and hence suffer from a limited information bandwidth. In this
\nchapter, we propose a novel concept of a high-bandwidth feedback interface
\nthat relies on advanced sensing and stimulation to convey a large amount
\nof information to the prosthesis user. The interface comprises an artificial
\nskin covering the prosthesis with a dense network of tactile sensors (taxels)
\nand a compact stimulation device delivering electrical current pulses through
\na matrix electrode with many conductive pads. The state-of-the-art in the
\ntwo technologies that are required for the implementation of the proposed
\nconcept are reviewed. This includes biomimetic e-skins that are suitable
\nfor the application in a wearable scenario, stimulation systems integrating
\na demultiplexing circuit to distribute electrical pulses, and flexible electrodes
\nthat can be produced with arbitrary shape, size, and distribution of conductive
\npads. Finally, the challenges in selecting feedback variables (raw signals
\nversus high-level features) and mapping of these variables into stimulation
\nparameters are addressed.