Reproduce “TacSpin: Tactile Tracking of In-Gripper Object Rotation on SO(3) via Axis-Magnitude Estimation” 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
Yuhao Wu, Linhao Song, Huan Yin
TacSpin estimates object rotation from two vision-based tactile sensors. It separately estimates each finger’s response direction and strength, refines the axis using boundary shear, then aligns and fuses estimates before accumulating rotation. It has lower mean checkpoint error than TacSE3 and TacSE3 + full-K across tested fixed-axis conditions. On mixed-axis sphere tracking, its mean trajectory error matches full-K, but endpoint error is higher.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.