Reproduce “TEAR: Task-Evidence Authority Routing for Robotic 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
Shicheng Zhang, Hui Liu, Haoliang Li
TEAR selects between remote teleoperation and a local policy using network measurements and visual task cues to predict handoffs. In ten simulation rollouts of orange-to-plate pick-and-place, it achieves 90% success, matching continuous teleoperation and exceeding the local policy by 30 percentage points. It uses teleoperation for 19.8% of control steps. The findings cover two related tasks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.