Learning-enhanced electronic skin for tactile sensing on deformable surface based on electrical impedance tomography

TL;DR

This work addresses the challenge of EIT-based tactile sensing on deformable surfaces where surface strains distort tactile readouts. It introduces VD2T, a dual-channel learning model that fuses EIT measurements with 3D deformation data to reconstruct tactile maps directly on deformed surfaces, using a pre-deformation reference without requiring real-time forward modeling. The approach is implemented on a hydrogel-based flexible e-skin and validated through multiphysics simulations and real-world experiments, achieving high correlation, PSNR, and low error metrics in simulation (CC 0.9660–0.9999, PSNR 28.72–55.53 dB, RIE 0.0107–0.0805) and DE ≈ 3.37% in experiments. The results demonstrate deformation-aware tactile reconstruction with potential to enhance soft-robotic tactile sensing and interaction on highly deformable surfaces.

Abstract

Electrical Impedance Tomography (EIT)-based tactile sensors offer cost-effective and scalable solutions for robotic sensing, especially promising for soft robots. However a major issue of EIT-based tactile sensors when applied in highly deformable objects is their performance degradation due to surface deformations. This limitation stems from their inherent sensitivity to strain, which is particularly exacerbated in soft bodies, thus requiring dedicated data interpretation to disentangle the parameter being measured and the signal deriving from shape changes. This has largely limited their practical implementations. This paper presents a machine learning-assisted tactile sensing approach to address this challenge by tracking surface deformations and segregating this contribution in the signal readout during tactile sensing. We first capture the deformations of the target object, followed by tactile reconstruction using a deep learning model specifically designed to process and fuse EIT data and deformation information. Validations using numerical simulations achieved high correlation coefficients (0.9660 - 0.9999), peak signal-to-noise ratios (28.7221 - 55.5264 dB) and low relative image errors (0.0107 - 0.0805). Experimental validations, using a hydrogel-based EIT e-skin under various deformation scenarios, further demonstrated the effectiveness of the proposed approach in real-world settings. The findings could underpin enhanced tactile interaction in soft and highly deformable robotic applications.

Figures (10)

• Figure 1: Schematic illustration of EIT-based tactile sensing.
• Figure 2: Fabrication process of the EIT-based tactile e-skin. (a) 3D printed mould. (b) Deploy electrodes on the mould. (c) Eco-flex was poured into the mould. (d) Cure at room temperature for 4 hours and release the mould. (e) The pre-gel solution of hydrogel was poured onto the cured silicone. (f) The hydrogel was polymerized by exposing it to UV light (365 nm) for 2 hours. (g) The fabricated EIT-based tactile sensor.
• Figure 3: The architecture of the proposed VD2T model.
• Figure 4: Tactile reconstruction on deformed surface. (a) Initial finite element model. (b) Tactile reconstruction based on the traditional method.
• Figure 5: Three types of touch patterns in the dataset. (a) One unit: Each unit is touched individually and sequentially. (b) Square-shaped area: Touch is applied to square areas ranging from 2 $\times$ 2 to 9 $\times$ 9 units sequentially. (c) Random 2 to 4 units.