Reproduce “ForceWAM: Decoupling Future and Response Learning from Demo and Play Data 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
Zijun Xu, Wendi Chen, Xinkai Wang, Jiachen Zou, Hongjin Chen, Hairuo Liu, Bokai Lin, Shihao Ma and 11 more
ForceWAM predicts future wrist-camera images and force/torque trajectories for a task, while learning action responses from successful demonstrations and exploratory play. It updates actions using measured wrench feedback between slower future updates. In force-regulated sliding, it achieves 85% success, versus 40% with task-only training and 20% for the strongest external baselines. Rates span 85% to 95% across four tested conditions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.