Infinigen-Sim: Procedural Generation of Articulated Simulation Assets

2025-05-15

AI Summary

To address low modeling efficiency, limited diversity, and poor cross-platform compatibility in articulated object simulation for robotics, this paper proposes the first end-to-end procedural asset generation framework tailored for articulated object segmentation, generalizable reinforcement learning, and sim-to-real transfer. The method leverages Blender to implement parametric articulated structure modeling—supporting joint constraints and physical plausibility—integrates differentiable rendering for automated semantic annotation, and establishes a unified export pipeline compatible with PyBullet, Isaac Gym, and MuJoCo. We develop dedicated generators for five common articulated object categories (e.g., cabinet doors, drawers, folding chairs). Experiments demonstrate that the generated assets significantly improve semantic segmentation accuracy (+12.3% mIoU), policy generalization across tasks (+18.7% success rate), and sim-to-real transfer performance (+24.1% success rate).

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Natural Language Processing: Generation

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

We introduce Infinigen-Sim, a toolkit which enables users to create diverse and realistic articulated object procedural generators. These tools are composed of high-level utilities for use creating articulated assets in Blender, as well as an export pipeline to integrate the resulting assets into common robotics simulators. We demonstrate our system by creating procedural generators for 5 common articulated object categories. Experiments show that assets sampled from these generators are useful for movable object segmentation, training generalizable reinforcement learning policies, and sim-to-real transfer of imitation learning policies.

Problem

• Procedural generation of diverse articulated objects
• Integration of assets into robotics simulators
• Enhancing sim-to-real transfer for learning policies

Innovation

• Procedural generation of articulated objects
• Blender utilities for asset creation
• Export pipeline for robotics simulators