Reproduce “Learning Low-Frequency Motion Control for Robust and Dynamic Robot Locomotion” 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
Siddhant Gangapurwala, Luigi Campanaro, Ioannis Havoutis
The authors test learned legged-robot controllers that update desired joint positions at low frequencies while a faster low-level tracker follows them. In simulation, blind 10-Hz and 200-Hz policies tolerate maximum added command delays of 90 ms and 20 ms, respectively. On ANYmal C, an 8-Hz blind policy tracks about 1.5 m/s. Yet 8-Hz reactivity trails 10-Hz on rough terrain.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.