TacTape: Real-Time High-Accuracy Tactile Fiducial System with Structured 3D Texture for Vision-Based Tactile Sensors
Meng Wang, Wanlin Li, Qiuxuan Chen, Yuzhe Huang, Hang Li, Kaspar Althoefer, Ziyuan Jiao, Yao Su, Hangxin Liu
AI summary
Problem
Vision-based tactile sensors struggle with localization on smooth or repetitive surfaces due to their small sensing area and lack of distinct surface features.
Approach
The authors developed TacTape, a flexible tape with a parameterized, automated 3D structured texture, paired with a lightweight geometric algorithm that decodes partial pattern observations to estimate contact position and orientation in real time.
Key results
- Sub-millimeter positional and sub-degree angular localization accuracy
- Over 99% decoding success rate from partial pattern observations
- Real-time processing at 30 Hz with under 3 ms computation per frame
- Successful dynamic posture compensation during human-robot tool handovers
Why it matters
Provides a scalable, easy-to-deploy solution that extends the utility of vision-based tactile sensors to smooth or unstructured surfaces for advanced robotic manipulation.
Abstract
Vision-based tactile sensors enable high-resolution tactile perception by capturing image-based contact data. How- ever, their utility in tactile localization is limited by their inher- ently small and local sensing area, as well as their dependence on distinct object surface features. We propose TacTape, a novel tactile fiducial system that enables accurate and efficient tactile localization by attaching textured tape to object surfaces. A lightweight algorithm allows real-time estimation of contact position and orientation from partially observed structured 3D textures. Experiments demonstrate that TacTape achieves sub-millimeter positional and sub-degree angular localization accuracy, and operates significantly faster than classic tactile mapping methods.