Reproduce “Foundation-Model-Guided Topology-Aware Semantic Risk Fields for Manipulation” 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
Giung Lee, Weihang Guo, Lydia E. Kavraki
The authors introduce a spatial semantic risk field for collision-constrained planning. It combines directional object-pair priors with geodesic distances and barrier attenuation, as a soft cost rather than a safety constraint. In laptop simulations, risk-aware CHOMP-style planning reduced mean reference-field risk from 0.580 to 0.408 and maximum risk from 0.999 to 0.691; results do not measure real-world damage.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.