Reproduce “Shape-Aware Neural Time Fields for Rigid-Body Motion and Manipulation 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
Yanming Zhou, Yuchen Liu, Ahmed H. Qureshi
Shape-Aware Neural Time Fields plans rigid-body motion by learning a reusable pose-to-pose cost-to-go for a fixed body and static environment. It supervises translation and rotation separately using geometric clearance targets, then guides sampling-based planning with goal progress and local motion difficulty. In the 3D benchmark, it achieved 92.9% success versus TDM’s 63.7%, within the tested settings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.