FEM-Validated Real-Time Modelling of Soft Tactile Deformation

Key points

• Accurate modelling of soft tactile sensors remains difficult because finite element methods (FEM) are computa- tionally expensive, while real-time simulators often lack fidelity for hyperelastic deformation.1
• We evaluate both forward prediction from marker deformation to contact localisation and inverse inference from contact configuration to marker deformation.1
• Models trained entirely on synthetic FEM data retain sub-millimetre accuracy on real measurements from the 2D tactile structure, demonstrating strong sim-to-real transfer within the tested planar setting.1
• We further integrate the learned model into a real-time visualisation pipeline, enabling rapid deformation prediction for virtual prototyping and simulator-based design exploration without extensive physical testing.1

Sources (1)

• [1]FEM-Validated Real-Time Modelling of Soft Tactile Deformation
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 12, 12:15 PM

  We present a data-efficient framework that combines limited real-world validation, high-fidelity FEM, and graph neural networks (GNNs) for accurate real-time prediction of two-dimensional tactile deformation in a planar soft tactile structure.

  Accurate modelling of soft tactile sensors remains difficult because finite element methods (FEM) are computa- tionally expensive, while real-time simulators often lack fidelity for hyperelastic deformation.

Extractive summary: sentences quoted from the sources.