aiqbench analysis
This research introduces USDCraft, a framework leveraging large language models to programmatically reconstruct articulated 3D assets from partial geometric data for robot simulation. By converting mesh inputs into metric textual descriptions, the system enables real-to-sim transfer without task-specific training, addressing critical gaps in handling incomplete or corrupted real-world objects. The approach demonstrates superior articulation recovery on benchmarks, facilitating more robust deployment of robotic manipulation policies in physical environments.
USDCraft: Geometrically Grounded Programmatic Modeling of Articulated 3D Assets for Simulation
arXiv:2610.11322v1 Announce Type: new Abstract: Geometrically faithful and functional articulated 3D assets are essential for real-to-sim robot manipulation, where policies trained in simulation must transfer to physical objects. Recent mesh-based methods learn to infer articulation from annotated 3D assets, but deployment remains challenging when real-world objects fall outside the training distribution or their meshes are incomplete or corrupted. To address these limitations, we formulate articulated asset reconstruction as programmatic modeling grounded in partial geometric evidence and introduce USDCraft, a framework in which a pretrained LLM writes and revises executable programs for simulation-ready articulated assets without task-specific training. We propose source geometry analysis, which converts the source mesh into a metric textual description that distinguishes observed surface from unknown space, and iterative geometric rechecking, which re-encodes each candidate in the same representation so that discrepancies point to program edits while unobserved regions remain open to completion. Visual feedback and physical authoring guidance complete the modeling process, which produces articulated USD assets with explicit physical properties that load into Isaac Sim without manual adjustment. Experiments demonstrate leading articulation recovery on two benchmarks and validate USDCraft's effectiveness for real-to-sim-to-real robot manipulation.
Abstract:Geometrically faithful and functional articulated 3D assets are essential for real-to-sim robot manipulation, where policies trained in simulation must transfer to physical objects. Recent mesh-based methods learn to infer articulation from annotated 3D assets, but deployment remains challenging when real-world objects fall outside the training distribution or their meshes are incomplete or corrupted. To address these limitations, we formulate articulated asset reconstruction as programmatic modeling grounded in partial geometric evidence and introduce USDCraft, a framework in which a pretrained LLM writes and revises executable programs for simulation-ready articulated assets without task-specific training. We propose source geometry analysis, which converts the source mesh into a metric textual description that distinguishes observed surface from unknown space, and iterative geometric rechecking, which re-encodes each candidate in the same representation so that discrepancies point to program edits while unobserved regions remain open to completion. Visual feedback and physical authoring guidance complete the modeling process, which produces articulated USD assets with explicit physical properties that load into Isaac Sim without manual adjustment. Experiments demonstrate leading articulation recovery on two benchmarks and validate USDCraft's effectiveness for real-to-sim-to-real robot manipulation.
Robotics (cs.RO)
arXiv:2610.11322 [cs.RO]
(or arXiv:2610.11322v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.11322 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 276 words
Related coverage
CoDimRecon: Agentic Reconstruction of Sim-Ready 3D Scenes with Deformable Curves, Surfaces, and Volumes
arXiv · Sep 30, 2026
The State of Simulation for Physical AI: An Overview
Hugging Face · Jul 21, 2026
PhySGA: A Physics-Space Grounded Agentic Framework for Generalizable and Interactive 3D Composition
arXiv · Oct 15, 2026
5 Steps to Create SimReady Assets for Robotics with Frontier AI Models
NVIDIA · Oct 8, 2026