Reproduce “Learn-To-Propagate: Distributionally Robust Motion Planning for Nonlinear Uncertain Systems” 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
Eduardo Figueiredo, Adrien Banse, Qi Heng Ho, Raphael Jungers, Morteza Lahijanian, Luca Laurenti
L2P-DR-Forest plans safe robot trajectories under unknown motion uncertainty by propagating data-learned sets of plausible distributions through nonlinear dynamics and refreshing them at promising restart points. It grows linked search trees, accepting steps only when worst-case obstacle risk meets threshold. In 20 3D AV Maze trials, it found plans in all, compared with 17/20 for ε-RandUP-RRT; these results are empirical.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.