Reproduce “Multi-Agent Reinforcement Learning of Co-Adaptive Hip Exoskeleton Control (MARL-Exo) for Metabolic Energy Reduction” 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
Yifei Yuan, Jakob Wolf, Ghaith Androwis, Xianlian Zhou
MARL-Exo trains simulated humans and an exoskeleton together, using a physiological muscle-energy model to select assistance. It transfers one controller without per-person tuning. In eight healthy adults walking on a treadmill at 1.25 m/s, assistance lowered net metabolic rate by 18.3% versus the passive device in all eight. Compared with walking without the device, six had lower cost, but the mean reduction was not statistically significant.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.