Reproduce “BiC-MPPI: Goal-Pursuing, Sampling-Based Bidirectional Rollout Clustering Path Integral for Trajectory Optimization” 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
Minchan Jung, Kwangki Kim
BiC-MPPI grows robot routes through obstacles by clustering samples rolled from the start and goal, joining nearby branches, and using the joined route to guide MPPI sampling. In 300 simulated quadrotor-landing trials, it achieved 96.7% success, versus 73.0% for Cluster-MPPI. Success required landing within 200 iterations without collision; this does not establish real-world flight performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.