Reproduce “Rapid Mean-Field Formation Transport with a Shared-Representation Neural Fokker-Planck Controller” 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
Zejian Zhou, Hao Xu
The authors introduce SR-NFPC, a shared representation that models swarm density and provides feedback control. During offline training, it learns density and control maps; online, each agent evaluates the fixed feedback map from its state and time. In five simulated 2D formation tasks, it achieves the lowest Wasserstein distance on all five, but a 1D planner has lower terminal error.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.