Reproduce “Learning Control Policies for Fall prevention and safety in bipedal 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
Visak Kumar
The dissertation presents reinforcement-learning methods for humanoid fall shaping and simulated assistive-device recovery. For unavoidable falls, a controller selects body contacts and motions; two hardware demonstrations replayed precomputed poses without feedback, reducing head acceleration from 5.7 to 2.9 G and 6.4 to 2.3 G. An error-aware policy expanded stability regions across five held-out simulated human models, not real device users.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.