Reproduce “Learning-Enhanced Observer for Sensorless Task-Null-Space Compliant 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
Nuo Chen, Ya-Jun Pan, Onur Toker
The authors present a force observer for a redundant robot. The learned correction uses task-related joint velocity, while null-space motion supports compliant control. In bowl-wiping tests with body disturbances, z-axis force-estimation RMSE fell from 2.016 N to 0.804 N versus the full-velocity observer, a 60.1% reduction in one disturbance window. The result is limited to the tested robot and conditions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.