Reproduce “Neuromuscular Reinforcement Learning to Actuate Human Limbs through FES” 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, Ali Shafti, A.Aldo Faisal
The authors develop a reinforcement-learning controller for functional electrical stimulation, using a recurrent network to summarize movement and stimulation history. They test it in simulated arm and cycling tasks and on healthy volunteers tracking elbow targets. In the 70°/20° task, PID has lower RMSE, 14.44° versus 17.31°, while RL is more consistent; the result does not show general accuracy superiority.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.