FlexiTac is a scalable, flexible tactile sensor designed to make touch sensing accessible for robotics. It integrates easily into diverse platforms and manipulation tasks. Our goal is to make tactile sensing simpler to build, easier to customize, and more practical to deploy.
FlexiTac is a flexible tactile sensing solution designed to make tactile sensing accessible to the robotics community. Built with scalability and adaptability in mind, FlexiTac can be integrated into a wide range of robotic platforms and manipulation tasks.
A short walkthrough of the tactile hardware fabrication and representative robotics applications.
Two sensor generations with different fabrication trade-offs, plus open-source resources for reproduction and deployment.
Fabrication time: ~3 mins
FPC-based · 2 mm grid · 32×12 resolution · $2.5 per sensor piece
Fabrication time: ~30 mins
3 mm grid · 16×16 resolution · $5 per sensor piece
Featured in 3D-ViTac (CoRL 2024) for fine-grained visuo-tactile manipulation.
Penalty-based tactile simulation for different simulation platforms.
A tactile simulation in Isaac Sim and Web-based interaction demo for data collection.
Featured in VT-Refine (CoRL 2025) for bimanual assembly via visuo-tactile simulation fine-tuning.
Representative examples showing how FlexiTac can be integrated into different robotic platforms.
We are excited to see FlexiTac reach labs, students, and researchers around the world.
Thank you for visiting and supporting open-source tactile sensing.
Selected research projects closely connected to FlexiTac and the broader area of visuo-tactile manipulation.
(* indicates equal contribution; + indicates equal advising)
3D-ViTac: Learning Fine-Grained Manipulation with Visuo-Tactile Sensing
Binghao Huang,
Yixuan Wang,
Xinyi Yang,
Yiyue Luo,
Yunzhu Li
Conference on Robot Learning (CoRL), 2024
[Webpage]
[Paper]
[Hardware Tutorial]
[Video]
VT-Refine: Learning Bimanual Assembly with
Visuo-Tactile Feedback via Simulation Fine-Tuning
Binghao Huang ,
Jie Xu,
Iretiayo Akinola,
Wei Yang,
Balakumar Sundaralingam,
Rowland O'Flaherty,
Dieter Fox,
Xiaolong Wang,
Arsalan Mousavian,
Yu-Wei Chao,
Yunzhu Li
Conference on Robot Learning (CoRL), 2025
Best Paper Award at IROS 2025 AHFHR Workshop [Link]
[Webpage]
[Paper]
[Video]
[Code]
Touch in the Wild:
Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper
Xinyue Zhu*,
Binghao Huang *,
Yunzhu Li
[Webpage]
[Paper]
[Video]
[Code]
Conference on Neural Information Processing Systems (NeurIPS), 2025
Best Demo Award at RSS 2025 Workshop on Robot Hardware-Aware Intelligence [Link]
Multi-Modal Manipulation via Multi-Modal Policy Consensus
Haonan Chen,
Jiaming Xu*,
Hongyu Chen*,
Kaiwen Hong,
Binghao Huang,
Chaoqi Liu,
Jiayuan Mao,
Yunzhu Li,
Yilun Du+, and
Katherine Driggs-Campbell+
[Webpage]
[Paper]
[Video]
[Code]
[Dataset]
[Audio]
[Blog]
[Deepwiki]
International Conference on Robotics & Automation (ICRA), 2026
Featured in Video Friday on
[IEEE Spectrum]
Tactile-Reactive Gripper with an Active Palm for Dexterous Manipulation
Yuhao Zhou,
Won Suk Lee,
Yan Gu,
Yu She
[Paper]
[Video]
Nature Portfolio Journal Robotics, 2026
Object Pose Estimation through Dexterous Touch
Amir-Hossein Shahidzadeh*,
Jiyue Zhu*,
Kezhou Chen,
Sha Yi,
Cornelia Fermüller,
Yiannis Aloimonos, and
Xiaolong Wang
[Webpage]
[Paper]
Learning Force-Regulated Manipulation with a Low-Cost Tactile-Force-Controlled Gripper
Xuhui Kang*,
Tongxuan Tian*,
Sung-Wook Lee,
Binghao Huang,
Yunzhu Li, and
Yen-Ling Kuo
[Webpage]
[Paper]
[Hardware Tutorial]
TacVLA: Contact-Aware Tactile Fusion for Robust Vision-Language-Action Manipulation
Kaidi Zhang*,
Heng Zhang*,
Zhengtong Xu,
Zhiyuan Zhang,
MD Rakibul Islam Prince,
Li Xiang,
Xiaojing Han,
Yuhao Zhou,
Arash Ajoudani, and
Yu She
[Webpage]
[Preprint]
Under Review
VTAP Gripper: Synergizing Fingertip Sensing and a Visuo-Tactile Active Palm for Dexterous In-Hand Manipulation
Yuhao Zhou,
Sheeraz Athar*,
Zhixian Hu*,
Binghao Huang,
Yunzhu Li,
Juan Wachs, and
Yu She
Under Review
Learning Tactile-Aware Quadrupedal Loco-Manipulation Policies
Pokuang Zhou,
Yuhao Zhou,
Quan Luu,
Seungho Han,
Heng Zhang,
Binghao Huang,
Yunzhu Li,
Arash Ajoudani,
Zhengtong Xu, and
Yu She
Under Review
Once you place an order, it typically takes around 1–2 weeks for the PCB or FPC pieces to be fabricated and shipped, depending on your location and the shipping method.
In our experience, the sensor shows good consistency over time and can often be used for around a year without significant degradation. Some teams have collected thousands of demonstrations without breaking the sensor.
We usually recommend ordering a few extra sensor pieces as backups. In our practice, replacing a damaged sensor with a new one typically does not noticeably affect performance, suggesting good consistency across different sensor instances.
