Reproduce “Towards AI-controlled FES-restoration of movements: Learning cycling stimulation pattern with reinforcement learning” 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
Nat Wannawas, A. Aldo Faisal
The method finds muscle-stimulation timing patterns for FES cycling by training reinforcement learning in a cyclist-matched model, then refining the pattern offline with recorded cycling data. In one healthy participant, average speed was 49.83 RPM with the model-derived pattern and 52.37 RPM after refinement in 30-second sessions; this proof of concept used very low resistance and has not been tested in people with paralysis.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.