Reproduce “Control Barrier Functions for Safety-Aware Latent-Space Motion Planning” 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

Chandrakiran Ravula, Siddhanth Ramesh

The authors replace a collision classifier with a learned control barrier function (CBF) for latent-space planning. The planner proposes a step, then corrects it along the barrier gradient if it violates the condition. On fixed obstacles, it achieves 75.6% success versus 63.6% for the best of 35 classifier settings, but general random-obstacle performance is lower and decoded trajectories have no formal safety guarantee.

Summary by alphaXiv.

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