Physics-Driven Learning Framework for Tomographic Tactile Sensing

TL;DR

This paper introduces PhyDNN, a physics-driven deep learning framework that embeds the EIT forward model into the training process, significantly improving tactile sensing accuracy and artifact reduction.

Contribution

The work presents a novel differentiable forward-operator network enabling physics-guided training of deep models for EIT-based tactile sensing.

Findings

Outperforms traditional methods like NOSER and TV in shape and pressure reconstruction.

Reduces artifacts and produces sharper boundaries in conductivity maps.

Demonstrates effectiveness through extensive simulations and real experiments.

Abstract

Electrical impedance tomography (EIT) provides an attractive solution for large-area tactile sensing due to its minimal wiring and shape flexibility, but its nonlinear inverse problem often leads to severe artifacts and inaccurate contact reconstruction. This work presents PhyDNN, a physics-driven deep reconstruction framework that embeds the EIT forward model directly into the learning objective. By jointly minimizing the discrepancy between predicted and ground-truth conductivity maps and enforcing consistency with the forward PDE, PhyDNN reduces the black-box nature of deep networks and improves both physical plausibility and generalization. To enable efficient backpropagation, we design a differentiable forward-operator network that accurately approximates the nonlinear EIT response, allowing fast physics-guided training. Extensive simulations and real tactile experiments on a 16-electrode soft sensor show that PhyDNN consistently outperforms NOSER, TV, and standard DNNs in reconstructing contact shape, location, and pressure distribution. PhyDNN yields fewer artifacts, sharper boundaries, and higher metric scores, demonstrating its effectiveness for high-quality tomographic tactile sensing.