Register or Log In

International Society for Industrial Process Tomography

12th World Congress on Industrial Process Tomography

Shape-constrained Electrical Impedance Tomography for Tactile Perception Under Complex Deformations

Sihao Teng1, Hao Fang1, Zhe Liu1, Jiabin Jia1, Yunjie Yang1

1 Institute for Imaging, Data and Communications (IDCOM), School of Engineering
The University of Edinburgh, Edinburgh, United Kingdom

*Email: y.yang@ed.ac.uk


ABSTRACT

Electrical Impedance Tomography (EIT) sensors offer significant potential for flexible tactile sensing due to their simplified wiring and adaptable mechanical properties. However, complex sensor deformations like twisting and stretching can severely degrade the accuracy of tactile reconstructions by causing sensitivity model mismatches. To address this challenge, we propose a Shape-Constrained flexible EIT (SC-fEIT) tactile reconstruction framework. Our approach integrates optical marker-based deformation tracking to dynamically recalibrate the sensitivity matrix. Additionally, we introduce a Sparse Shallow Image Prior (SSIP) reconstruction method, utilizing a sparsely connected multilayer perceptron (Sparse-MLP) to achieve robust tactile reconstruction under limited and noisy measurements. Handcrafted regularizations are utilized to stabilize the inversion process. Our approach is validated through simulations and experiments on real-world tactile phantoms, demonstrating reliable reconstruction of conductivity distributions and accurate localization of tactile stimuli despite significant sensor deformation. The proposed framework provides a robust solution for tactile perception using flexible EIT sensors under complex deformation conditions.


Keywords: Electrical Impedance Tomography, Flexible sensor, Tactile perception

Industrial Application: General

Sign-in to access the full text