Reproduce “PRIMO: Prior-Informed Odometry from Human-Motion Tracking for Humanoid Robots” 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
Xu Han, Angsong Li, Shaopeng Zhang, Enyu Li, Peiwen Lin, Chuang Wang, Yuan Zhuang, Haiyu Lan
PRIMO trains humanoid proprioceptive odometry using simulated robots tracking retargeted human motions, rather than relying on a deployment controller’s rollouts. Its estimator combines sensor histories with physics- and symmetry-informed structure. On one humanoid, it has the lowest mean error in all six real-robot domain–metric comparisons against four baselines, with reductions of 31.6% to 61.7%. Tests include ten dynamic trajectories.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.