Reproduce “FOMI: A Force-Feedback Manipulation Interface for Teaching Contact-Rich Manipulation” 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
Yishu Shen, Yiming Luo, Ruilin Wang, Bowen Yang, Shuai Yang, Fangyi Xu, Yujun Shen, Fu Zhang and 2 more
FOMI helps people collect robot-free manipulation demonstrations through a scissor-shaped interface that transmits gripper resistance mechanically to supported handles and maps motion to robot commands for teleoperation. Operators collected 17.8 successful demonstrations per 10-minute session, versus 13.8 with UMI. The exploratory comparison involved three novice operators; policies trained on FOMI data received no force or tactile inputs.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.