Website | Getting Started | Making Your First Skin | Tips and Tricks | More Modalities
Procedural skins are a new class of artificial skins for robotic applications designed to be form-fitting and highly customizable to individual robots and use-cases. Procedural skins utilize a CAD model of a robot to automatically generate sensors with directely tunable parameters such as sensor density and sensing coverage are directly tunable.
Procedural skins are designed in Blender, then can be optimized or deployed in simulation through Isaac Sim. The procedural skins developed here use RC delay capacitive sensing and are fabricated using multi-material 3D prinitng. Please refer to the original paper for more information on fabrication.
No prerequisite experience with Blender is needed to get started and design your first skin unit. All you need is a CAD model that you plan to cover with skin, then drag and drop the desired components onto the model from the asset browser.
Detailed instructions on how to design your first skin unit can be found in the Blender Procedural Generation Add-on README
Detailed instructions on how to import your skin unit to Isaac Sim can be found in the Contact Extension README
The procedural design is built on Blender's geometry node system. You can edit the designs by opening a geometry nodes window and navigating through the premade tabs.
This repository is under a Creative Commons Attribution-NonCommercial 4.0 International License, which allows for unrestricted sharing and modifications with attribution, but prohibits commercial use.
If you find our paper or codebase helpful, please consider citing:
@inproceedings{kohlbrenner2025gentact,
title={Design, Mapping, and Contact Anticipation with 3D-printed Whole-Body Tactile and Proximity Sensors},
author={Kohlbrenner, Carson and Soukhovei, Anna and Escobedo, Caleb and Nechyporenko, Nataliya and Roncone, Alessandro},
booktitle={2026 IEEE International Conference on Robotics and Automation (ICRA)},
year={2026}
}
@inproceedings{kohlbrenner2025gentact,
title={GenTact Toolbox: A Computational Design Pipeline to Procedurally Generate Context-Driven 3D Printed Whole-Body Artificial Skins},
author={Kohlbrenner, Carson and Escobedo, Caleb and Bae, S Sandra and Dickhans, Alexander and Roncone, Alessandro},
booktitle={2025 IEEE International Conference on Robotics and Automation (ICRA)},
year={2025}
}