Research Analyzer
← Back ICRA 2026

Design, Mapping, and Contact Anticipation with 3D-Printed Whole-Body Tactile and Proximity Sensors

Carson Kohlbrenner, Anna Soukhovei, Caleb Escobedo, Nataliya Nechyporenko, Alessandro Roncone

PDF

AI summary

Key figure (auto-extracted from paper)
A fully 3D-printed whole-body artificial skin combined with a data-driven machine learning framework enables robots to anticipate contact by mapping their effective proximity sensing volume.
artificial skin proximity sensing 3D printing perisensory space contact anticipation machine learning

Problem

Robots operating in dynamic environments lack scalable, whole-body artificial skins that can sense proximity, and existing methods struggle to map the surrounding space these sensors can effectively monitor for contact anticipation.

Approach

The authors procedurally generate and 3D-print modular skin units with embedded capacitive sensors, then use an ensemble of machine learning models trained on hover trajectories to map the Perisensory Space and predict nearby object positions without relying on fixed sensor geometry assumptions.

Key results

  • GenTact-Prox: a modular, procedurally generated 3D-printed whole-body artificial skin with integrated tactile and proximity sensing
  • A data-driven machine learning ensemble that maps the Perisensory Space and predicts object position uncertainty from arbitrary sensor layouts
  • Experimental validation on a Franka Research 3 robot demonstrating detection ranges up to 18 cm and successful online contact prediction
  • Open-sourced procedural design pipeline and printable sensor files for scalable, low-cost artificial skin fabrication

Why it matters

Lowers the barrier to creating interactive robots with anticipatory spatial awareness for safe human-robot collaboration and dynamic environment navigation.

Abstract

Robots operating in dynamic and shared environ- ments benefit from anticipating contact before it occurs. We present GenTact-Prox, a fully 3D-printed artificial skin that integrates tactile and proximity sensing for contact detection and anticipation. The artificial skin platform is modular in design, procedurally generated to fit any robot morphology, and can cover the whole body of a robot. The skin achieved detection ranges of up to 18 cm during evaluation. To characterize how robots perceive nearby space through this skin, we introduce a data-driven framework for mapping the Perisensory Space— the body-centric volume of space around the robot where sensors provide actionable information for contact anticipation. We demonstrate this approach on a Franka Research 3 robot equipped with five GenTact-Prox units, enabling online object- aware operation and contact prediction.

Index terms

Touch in HRI Bioinspired Robot Learning Multi-Modal Perception for HRI

Related papers