Reproduce “Force-Aware Residual DAgger via Trajectory Editing for Precision Insertion with Impedance Control” 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
Yiou Huang, Ning Ma, Weichu Zhao, Zinuo Liu, Jun Sun, Qiufeng Wang, Yaran Chen
TER-DAgger improves robot insertion by using force prediction errors to request human corrections and learn residual adjustments to its existing motion. It aligns each correction with the base trajectory and executes commands through Cartesian impedance control. Across five simulated and real-world insertion settings, it averages 77.2% success versus 40.0% for fine-tuning, but succeeds in only 22% of real-world USB trials.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.