Reproduce “ForceBody: Force-Paired Parametric Body Motion with Torque Uncertainty” 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

Joonwoo Kwon, Yufei Zhang, Xiaoming Liu, Zijun Cui

ForceBody pairs measured ground-reaction forces and inverse-dynamics torque estimates with motion refit to SKEL, a parametric body model with anatomically grounded joints. It provides torque uncertainty estimates based on joint-angle noise. The dataset contains 10,386 trials. In a Transformer experiment, uncertainty-weighted training reduced all-joint torque MAE from 0.107 to 0.101 Nm/kg, a finding limited to that experiment.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.