Reproduce “Whisker-Inspired Tactile Sensing: A Sim2Real Approach for Precise Underwater Contact Tracking” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Hao Li, Chengyi Xing, Saad Khan, Miaoya Zhong, Mark R. Cutkosky
A whisker-inspired tactile sensing system, coupled with a sim-to-real deep learning framework, enables precise underwater contact tracking. This approach achieves less than 2 mm accuracy in real-world underwater experiments on diverse objects, critically operating without explicit robot proprioceptive data.
Key insights
- Whisker-inspired tactile sensors, incorporating robust FBG technology, provide an effective modality for underwater perception in low-visibility conditions.
- A sim-to-real learning framework, including a real-to-sim calibration, can effectively train deep learning models for tactile sensing, mitigating the need for extensive real-world data collection.
- Accurate contact localization can be achieved from whisker base torque signals alone, eliminating the dependency on explicit robot proprioception.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.