Reproduce “FTP-1: A Generalist Foundation Tactile Policy Across Tactile Sensors for 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
Chengbo Yuan, Zicheng Zhang, Mingjie Zhou, Wendi Chen, Yi Wang, Zhuoyang Liu, Dantong Niu, Shuo Wang and 9 more
FTP-1 tackles sensor incompatibility by mapping tactile images, arrays, and force and state readings to tokens for functional hand regions, then processing them with a shared tactile expert alongside vision, language, and robot state. Pretrained on about 3,000 hours from 26 sources and 21 sensors, it averages 46.6% success after fine-tuning on demonstrations for two unseen hardware setups, versus 15.0% without pretraining.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.