Reproduce “MPPI-IPDDP: Hybrid Method of Collision-Free Smooth Trajectory Generation for Autonomous Robots” 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

Min-Gyeom Kim, Minchan Jung, JunGee Hong, Kwang-Ki K. Kim

MPPI-IPDDP combines sampling-based route search with sampled collision-free corridors and constrained trajectory optimization. It first finds a coarse route, then refines it within safe regions while respecting robot dynamics. In 300 BARN scenarios, it achieved 95.7% success and mean squared curvature of 0.000139, versus 0.002528 for vanilla MPPI, indicating smoother paths in this test, but not highest success rate.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.