A Machine Learning Approach to Contact Localization in Variable Density Three-Dimensional Tactile Artificial Skin

TL;DR

This paper presents a neural network-based method for contact localization on 3D artificial skin with non-uniform sensor distribution, achieving accurate touch point estimation despite complex geometry and sensor arrangement.

Contribution

It introduces a neural network approach for contact localization on non-uniform, semi-conical 3D tactile surfaces, addressing limitations of flat, uniform sensor models.

Findings

Localization error of 5.7 ± 3.0 mm achieved

Method works on complex 3D geometries

Robust to sensor distribution variability

Abstract

Estimating the location of contact is a primary function of artificial tactile sensing apparatuses that perceive the environment through touch. Existing contact localization methods use flat geometry and uniform sensor distributions as a simplifying assumption, limiting their ability to be used on 3D surfaces with variable density sensing arrays. This paper studies contact localization on an artificial skin embedded with mutual capacitance tactile sensors, arranged non-uniformly in an unknown distribution along a semi-conical 3D geometry. A fully connected neural network is trained to localize the touching points on the embedded tactile sensors. The studied online model achieves a localization error of $5.7 \pm 3.0$ mm. This research contributes a versatile tool and robust solution for contact localization that is ambiguous in shape and internal sensor distribution.