NVIDIA 用 AI 智能体为机器人仿真准备 3D 场景
原标题：How to Use AI Agents to Prepare 3D Scenes for Simulation
AI 摘要
NVIDIA 技术博客介绍了一种用 AI 智能体为机器人仿真准备 3D 场景的工作流。该流程以 Codex（由 OpenAI GPT-6 Astra 驱动）作为主协调智能体，通过 NVIDIA NemoClaw 部署基于 Hermes 等框架的专用子智能体，调用 NVIDIA Omniverse Libraries（OpenUSD、ovphysx、ovrtx、SimReady 验证）对 Blender 场景进行语义标注、物理属性配置、碰撞体生成和预检渲染，最终输出可用于 Isaac Sim 或 Isaac Lab 的仿真就绪 OpenUSD 世界。文章指出机器人训练的真正瓶颈常在于缺少仿真就绪环境，并强调安全机械性问题可自动修复、依赖开发者意图的模糊决策需升级给人工审核。
正文节选
Agentic AI workflows can be used to prepare and validate digital twins for physical AI systems. Agents can inspect 3D scenes, author simulation-relevant data in OpenUSD, add physics properties, render preflight views, and validate the result against simulation-ready (SimReady) requirements. This workflow follows that process from a scene in Blender to a simulation-ready OpenUSD handoff for NVIDIA Isaac Sim or NVIDIA Isaac Lab. In practice, however, if you’re building agents for robotics, the wor