推出 SimReady Central：面向机器人与 AI 工厂的仿真就绪 3D 资产开放库

Back to Articles

构建机器人仿真和高保真 AI 工厂数字孪生，所需的不只是 3D 模型。开发者通常还需要添加碰撞几何体、质量和摩擦等属性。SimReady 资产是预先配置好仿真工作流所需属性的 3D 模型，可减少这部分手动工作。Building robotics simulations and high-fidelity AI factory digital twins takes more than 3D models. Developers often need to add properties such as collision geometry, mass, and friction. SimReady assets are 3D models that come prepared with the properties required for simulation workflows, reducing that manual work.

SimReady 中央库汇集了 Hugging Face 上 2,000 多个可用于仿真的 (SimReady) OpenUSD 资产，为开发者构建物理 AI 训练环境和 AI 工厂数字孪生提供免费、无门槛的起点。该库包含各种基础设施和设备资产，从机械臂和夹爪，到仓库设备、货架和冷却基础设施。开发者可以查看每项资产支持的功能，然后将来自不同供应商的资产组合到同一个场景中。SimReady Central Library brings together more than 2,000 simulation-ready (SimReady) OpenUSD assets on Hugging Face, giving developers a free, ungated starting point for physical AI training environments and AI factory digital twins. The library includes a variety of infrastructure and equipment assets from robot arms and grippers to warehouse equipment, racks, and cooling infrastructure. Developers can check which capabilities each asset supports, then combine assets from different suppliers into a scene.

设备制造商（OEM）现在可以根据 SimReady 要求准备并验证资产，并使用 SimReady Central Library 触达开发者，让其终端客户能够更快地测试和模拟各种场景，加快决策。这使各行业能够围绕自身产品构建仿真，同时管理自己的 Hugging Face 数据集。Equipment manufacturers (OEMs) can now prepare and validate assets against SimReady requirements and use SimReady Central Library to reach developers and allow their end customers to test and simulate scenarios faster and accelerate decision making. This enables industries to build simulations around their products while managing their own Hugging Face datasets.

我们将介绍资产具备 SimReady 标准的条件，以及 OEM 如何为 SimReady Central Library 做出贡献；随后展示开发者如何利用这些资产构建机器人抓取与放置场景，并比较 AI 工厂布局。We’ll cover what makes an asset SimReady and how OEMs contribute to SimReady Central Library, then show how developers can use those assets to build a robotic pick-and-place scenario and compare AI factory layouts.

什么样的资产才算是 SimReady？ What makes an asset SimReady?

SimReady 资产是经过准备、具备仿真工作流所需属性的 OpenUSD 资产。详细的 3D 模型（例如 CAD）是一个有用的起点。仿真还需要额外的信息，以说明应如何解读和使用该对象。SimReady assets are OpenUSD assets prepared with the properties required for simulation workflows. A detailed 3D model, such as CAD, is a useful starting point. Simulation requires additional information about how that object should be interpreted and used.

对于机器人操作任务，这可能包括物体的质量、碰撞几何形状和摩擦特性。对于 AI 工厂工作流，这可能包括设备几何形状，以及应用所需的电气或热学元数据。For a robot manipulation task, that might include the object’s mass, collision geometry, and friction properties. For AI factory workflows, it might include equipment geometry and the electrical or thermal metadata required by the application.

OpenUSD 是一个开放、可扩展的框架，用于描述和组合 3D 场景。使用这些场景训练和评估自主机器，还需要有关物理行为、碰撞和传感器交互的信息。OpenUSD is an open, extensible framework for describing and composing 3D scenes. Using those scenes to train and evaluate autonomous machines also requires information about physical behavior, collisions, and sensor interactions.

SimReady Foundation 提供规范、配置文件和验证工具，明确规定这些要求。配置文件定义了资产针对特定使用场景所需功能和能力的逻辑分组。SimReady Foundation provides specifications, profiles, and validation tools that make these requirements explicit. A profile defines the logical grouping of features and capabilities an asset needs for a particular use case.

图 1。SimReady Foundation 为特定仿真用例定义资产表示要求，而开发者库则帮助根据这些要求准备和验证 OpenUSD 资产。Figure 1. SimReady Foundation defines asset representation requirements for specific simulation use cases, while developer libraries help prepare and validate OpenUSD assets against those requirements.

根据资产和配置文件的不同，这些功能可能包括：Depending on the asset and profile, those features can include:

有关各仿真用例对应的要求，请参阅 SimReady Foundation 要求和配置文件。开发者应从 SimReady Central 中选择配置文件符合其工作流需求的资产，并将其组合到 OpenUSD 场景中。For the requirements associated with each simulation use case, see the SimReady Foundation requirements and profiles. Developers should select assets from SimReady Central whose profiles match their workflow needs and compose them into an OpenUSD scene.

为什么要将 OEM 资产发布到 SimReady Central？ Why publish OEM assets to SimReady Central?

OEM 可以将其设备的 CAD 模型转化为 SimReady 资产，供开发者围绕其产品构建仿真。SimReady Central Library 让这些资产可在 Hugging Face 上被发现，帮助制造商触达正在构建机器人仿真和 AI 工厂数字孪生的团队。OEMs can turn their CAD equipment models into SimReady assets for developers building simulations around their products. SimReady Central Library makes those assets discoverable on Hugging Face, helping manufacturers reach teams building robotics simulations and AI factory digital twins.

对于 Vertiv 等 AI 工厂 OEM 厂商，该资源库提供了一种途径，可分发仓库设备机架、服务器机架、冷却设备以及 AI 工厂数字孪生所需的配套基础设施等资产。开发者和前沿智能体可以将这些资产集成到 AI 工厂数字孪生中，并复用其 SimReady 配置文件中指定的几何形状和属性。For AI factory OEMs like Vertiv, the library provides a path to distribute assets such as warehouse equipment racks, server racks, cooling equipment, and supporting infrastructure for AI factory digital twins. Developers and frontier agents can incorporate these assets into AI factory digital twins reusing the geometry and properties specified by their SimReady profiles.

对于 Comau、FANUC、Franka Robotics、Robotiq、Standard Bots 和 Universal Robots 等机器人 OEM 厂商，该库让构建训练和评估环境的团队能够发现机械臂和夹爪。开发者可以将这些资产与工业设备结合，构建仓库和工厂场景。For robotics OEMs like Comau, FANUC, Franka Robotics, Robotiq, Standard Bots, and Universal Robots, the library makes robot arms and grippers discoverable by teams building training and evaluation environments. Developers can combine these assets with industrial equipment to build warehouse and factory scenarios.

对于 Palatial 这类仿真资产提供商，其资源库包含与 NVIDIA Robotics 合作发布的 NVIDIA Robotics 1K 数据集，以及根据文本、图像、扫描件、数据表和 CAD 生成的复杂关节式资产。开发者可以构建用于机器人训练和评估的资产与场景。For simulation asset providers like Palatial the library features the NVIDIA Robotics 1K Dataset released in collaboration with NVIDIA Robotics, alongside complex articulated assets generated from text, images, scans, datasheets, and CAD. Developers can build assets and scenes for robot training and evaluation.

对于像 LG Electronics（LGE）这样的 OEM 厂商，他们在开发家用机器人的同时，也研发机器人会与之协同工作的家电产品。该资源库为共享冰箱、洗碗机等资源提供了途径。开发者可以利用基于 LGE 在家电和机器人领域经验打造的资源，构建用于机器人训练、评估和合成数据生成的家庭仿真环境。For OEMs like LG Electronics (LGE) that develop robots for the home alongside the appliances they work with, the library provides a path to share assets such as refrigerators and dishwashers. Developers can use assets informed by LGE’s experience in both appliances and robotics to build household simulation environments for robot training, evaluation, and synthetic data generation.

对于 ForgeCAD 等 AI 辅助 CAD 服务提供商而言，该素材库提供了带有关节结构且纹理质量出色的机器人道具，开发者可利用这些道具构建用于机器人感知与操作训练的环境，并生成合成数据。For AI-assisted CAD providers like ForgeCAD, the library makes articulated robotics props with high-quality textures available, which developers can use to build environments for robot perception and manipulation training, as well as synthetic data generation.

为了参与贡献，OEM 可以使用 SimReady Central 提交技能，根据相关 SimReady 配置文件验证资产、打包资产，并提交以纳入资源库。To contribute, OEMs can use the SimReady Central submission skill to validate assets against the relevant SimReady profiles, package them, and submit them for inclusion in the library.

图 2. OEM 根据 SimReady 配置文件准备并验证 OpenUSD 资产，然后将资产提交至 SimReady Central。开发者和 AI 智能体可在库中查找资产，并使用这些资产构建仿真环境。Figure 2. OEMs prepare and validate OpenUSD assets against SimReady profiles and then submit the asset to SimReady Central. Developers and AI agents discover assets in the library and use them to build simulation environments.

使用 SimReady 资产构建仿真环境 Build simulation environments with SimReady assets

SimReady Central 为机器人和工业领域的开发者提供可复用的起点，助其构建机器人工作单元、工业环境和 AI 工厂数字孪生。开发者和 AI 智能体可以在 SimReady Central 中查找符合其工作流所需配置的资产，并将这些资产组合成 OpenUSD 环境。SimReady Central gives robotics and industrial developers a reusable starting point for robot workcells, industrial environments, and AI factory digital twins. Developers and AI agents can find assets in SimReady Central that meet the profiles required by their workflows and compose them into OpenUSD environments.

开发者可以使用 SimReady 资产在训练环境中布置物体和障碍物，然后改变它们的位置来评估机器人的行为。这些环境可支持 Microduck 等项目。Microduck 是 Hugging Face 和 Pollen Robotics 推出的开源机器人。Developers can use SimReady assets to populate training environments with objects and obstacles, then vary their placement to evaluate robot behaviors. These environments can support projects such as Microduck, an open-source robot from Hugging Face and Pollen Robotics.

图 3. 资产数据集通过 Hugging Face 上的 SimReady Central 连接，并利用其 USD 搜索服务支持 Isaac 工作流、SimReady Central 库搜索和 AI 代理工作流。Figure 3. Asset datasets connect through SimReady Central on Hugging Face, using its USD search service to support Isaac workflows, SimReady Central library search, and AI agent workflows.

NVIDIA USD Search 为 SimReady Central Library 提供自然语言搜索功能，帮助开发者和 AI 智能体在参与的数据集中查找资产。Hugging Face 也已将 USD Search 集成到其平台搜索中，并索引 USD 属性和仿真元数据，帮助智能体发现适用于其物理 AI 仿真任务的 SimReady 资产。NVIDIA USD Search powers natural-language search in SimReady Central Library, helping developers and AI agents find assets across participating datasets. Hugging Face has also integrated USD Search into their platform search with indexed USD properties and simulation metadata, helping agents discover SimReady assets suited to their physical AI simulation tasks.

AI 辅助创作提供了另一种利用这些资产进行构建的方式。开发者可以将 OpenAI 的 GPT-6 Astra 等前沿模型与资产搜索和场景创作工具结合起来，使用 SimReady Central Library 构建环境。根据任务需求和空间限制，Astra 可以帮助查找合适的资产、编写组装场景的代码，并创建不同的场景变体。AI-assisted authoring adds another way to build with those assets. Developers can use frontier models like OpenAI’s GPT-6 Astra combined with asset search and scene-authoring tools to build environments using SimReady Central Library. Given task requirements and spatial constraints, Astra can help find suitable assets, write code to assemble a scene, and create scenario variations.

原始设备制造商也可以使用 Astra，根据 SimReady Foundation 的要求准备和优化 OpenUSD 资产，然后再对其进行验证并提交至资源库。OEMs can also use Astra to help prepare and refine OpenUSD assets against SimReady Foundation requirements before validating and submitting them to the library.

构建一个机器人抓取与放置场景 Build a robotic pick-and-place scenario

例如，开发者在构建机器人抓取与放置场景时，可以从以下提示词开始：For example, a developer building a robotic pick-and-place scenario could start with this prompt:

使用 Hugging Face 上 SimReady Central 的 SimReady 资产，在 Isaac Sim 中使用 PhysX 创建一个基础的仓库取放场景。通过 nvidia-simready-central.hf.space 查找一款 FANUC 机械臂、一款兼容的 Robotiq 夹爪以及合适的仓库资产。生成用于组装场景和配置取放任务的 Python 代码。渲染一个 .mov 预览，供我检查结果。请提供脚本、资产版本和运行时版本，以便我复现该场景。

Using SimReady assets from SimReady Central on Hugging Face, create a basic warehouse pick-and-place scenario in Isaac Sim using PhysX.

Use the nvidia-simready-central.hf.space to find a FANUC robotic arm, a compatible Robotiq gripper, and suitable warehouse assets. Generate the Python code to assemble the scene and configure the pick-and-place task. Render a .mov preview so I can review the results.

Include the scripts, asset versions, and runtime versions so I can reproduce the scenario.

在此工作流程中，我们期望 Astra 执行以下操作：In this workflow, we expect Astra to do the following:

• 调用 SimReady Search API，查找在 Hugging Face 上分发的相关资产。
  Query the SimReady Search API to find relevant assets distributed on Hugging Face
• 将它们作为引用/载荷引入 OpenUSD 场景”】【
  Reference/Payload them into an OpenUSD stage
• 生成用于组装工作单元的 Python 代码
  Generate Python code to assemble the workcell
• 配置拾取与放置任务
  Configure the Pick-and-place task
• 渲染预览供用户审核。
  Render a preview for user review

图 4. 由 SimReady 资产组装而成的机器人取放工作单元，展示了 AI 辅助场景创作如何将开发者需求转化为仿真场景，以便在 Isaac Sim 中进行审核和完善。Figure 4. A robotic pick-and-place workcell assembled from SimReady assets illustrates how AI-assisted scene authoring can turn developer requirements into a simulation scenario for review and refinement in Isaac Sim.

比较 AI 工厂布局 Compare AI factory layouts

同样的方法也可以支持独立软件供应商（ISV）构建 AI 工厂数字孪生。任务重点从机器人交互转向设备布局：在规定的空间限制内布置机架、冷却设备和配套基础设施。The same approach can support an ISV (independent software vendor) building an AI factory digital twin. The task shifts from robot interactions to equipment placement: arranging racks, cooling equipment, and supporting infrastructure within specified spatial constraints.

使用 Hugging Face 上 SimReady Central 的 SimReady 资产，创建两种备选的 AI 工厂布局，包含机架和冷却设备。请使用提供的设备尺寸和通道净空要求。检查设备是否存在重叠或违反净空要求的情况，并报告发现的问题。将每种布局保存为 OpenUSD 场景。请一并提供场景构建脚本、资产版本和运行时版本，以便我复现并比较结果。

Using SimReady assets from SimReady Central on Hugging Face, create two alternative AI factory layouts with racks and cooling equipment.

Use the supplied equipment dimensions and aisle-clearance requirements. Check for equipment overlaps and clearance violations, and report any issues.

Save each layout as an OpenUSD scene. Include the scene-building scripts, asset versions, and runtime versions so I can reproduce and compare the results.

在这一工作流中，Astra 会将选定的 SimReady 资产组装成两种 OpenUSD 布局，随后生成脚本，根据提供的要求检查设备是否重叠以及通道净空是否符合要求。开发人员随后可以在仿真应用程序中查看这些布局，比较设备摆放位置和通行路径，并在优化首选配置之前检查所有标记出的问题。In this workflow, Astra would assemble the selected SimReady assets into two OpenUSD layouts, followed up by generating scripts to check equipment overlaps and aisle clearances against the supplied requirements. Developers could then inspect the layouts in their simulation application, compare equipment placement and access paths, and review any flagged issues before refining the preferred configuration.

图 5. 由 SimReady 资产组装而成的 AI 工厂布局，展示了 AI 辅助场景创作如何支持设备放置、通道净空检查以及在 OpenUSD 中优化布局。Figure 5. An AI factory layout assembled from SimReady assets illustrates how AI-assisted scene authoring supports equipment placement, aisle-clearance checks, and layout refinement in OpenUSD.

在这两个示例中，开发者都可以通过几种方式扩展其环境：In either example, developers can extend their environments in a few ways:

• 复用环境：将布局和任务特定的更改分别保存在独立的 OpenUSD 图层中，这样同一组源资产就能支持多个实验。
  Reuse the environment: Keep layout and task-specific changes in separate OpenUSD layers so the same source assets can support multiple experiments.
• 变换场景：从库中引入其他资产，并调整对象摆放位置、设备配置或周边基础设施，以评估不同条件。
  Vary the scenarios: Introduce additional assets from the library and vary object placements, equipment configurations, or surrounding infrastructure to evaluate different conditions.
• 构建特定于应用的测试：对于机器人，需在 Isaac Lab 中配置训练和评估任务。对于 AI 工厂，需将场景连接到设备数据和仿真工具，以开展配电或冷却性能等分析。
  Build application-specific tests: For robotics, configure training and evaluation tasks in Isaac Lab. For AI factories, connect the scene to the equipment data and simulation tools needed for analyses such as power distribution or cooling performance.
• 将经过验证的布局应用于运营：对于 AI 工厂，将场景连接到实时 BMS 和 IT 遥测数据，使数字孪生与 AI 工厂设施保持同步。
  Carry validated layouts into operations: For AI factories, connect the scene to live BMS and IT telemetry to keep the digital twin synchronized with an AI factory facility.

助力塑造下一波资产娱乐主管 Help shape the next wave of assets

SimReady Central 的路线图包括引入更多 OEM 合作伙伴并不断扩充资产库，让开发者能够利用可复用内容组装更多设备和环境。SimReady Central’s roadmap includes more OEM partners and a growing asset library, expanding the equipment and environments developers can assemble from reusable content.

社区反馈将帮助我们确定哪些新增内容最重要。请在评论中告诉我们你需要哪些资源、使用哪些运行时，以及在哪些地方遇到了集成缺口。Community feedback will help identify which additions matter most. Tell us in the comments which assets you need, which runtimes you use, and where you encounter integration gaps.

按照资产提交指南提交 OEM 资产，以便将其纳入 SimReady Central。Submit OEM assets for inclusion in SimReady Central using the asset submission guide.

OEM 的贡献可以扩充可用内容，而开发者示例则可以展示资产在不同应用中的用法。欢迎在评论区分享你构建的示例；如果发现错误，或有关于 SimReady 需求的功能请求，请在 SimReady Foundation 中提交 issue。OEM contributions can broaden the available content, while developer examples can show how assets work in different applications. Share examples of what you build in the comments and open an issue in SimReady Foundation for bugs or SimReady requirements feature requests.

开始在 Hugging Face 🤗 上构建 Start building on Hugging Face 🤗

从 SimReady Central Library 中选择一个资产，用于你的下一个机器人仿真或 AI 工厂数字孪生项目。使用以下资源查找资产、构建场景或贡献设备模型。Choose an asset from SimReady Central Library for your next robotics simulation or AI factory digital twin. Use the resources below to find assets, build a scene, or contribute equipment models.

发现资产 Discover assets

• Find assets for your task: SimReady Central Library on Hugging Face
• 使用自然语言搜索：通过由 USD Search 提供支持的 SimReady Central 搜索端点，描述您需要的资产。
  Search with natural language: Describe the assets you need through the SimReady Central search endpoint powered by USD Search.
• 构建你自己的资产浏览器：探索 SimReady Central GitHub 仓库中从索引到资源库的技能。
  Build your own asset browser: Explore the index-to-library skills in the SimReady Central GitHub repository.

构建一个场景 Build a scene

• 学习如何组装 OpenUSD 场景：沿着免费的 OpenUSD 学习路径进行学习。
  Learn to assemble OpenUSD scenes: Follow the free OpenUSD learning path.
• 提升你的机器人技术和数字孪生技能：探索免费的物理 AI 课程。
  Develop your robotics and digital twin skills: Explore the free physical AI courses.
• 构建机器人仿真：从 Isaac Sim 和 Isaac Lab 入手。
  Build robotics simulations: Start with the Isaac Sim and Isaac Lab.
• 建设 AI 工厂：从智能体辅助的数字孪生工作流开始。
  Build AI factories: Start with an agent-assisted digital twin workflow.

准备并贡献内容 Prepare and contribute content

• 启动 OEM 贡献流程：按照入门指南操作，并在 SimReady Central Library 文档中查找常见问题解答。
  Start the OEM contribution process: Follow the onboarding guide and find FAQ answers in the SimReady Central Library documentation.
• 选择资产的要求：查看 SimReady Foundation 中的配置文件。
  Choose the requirements for your asset: Review the profiles in SimReady Foundation.
• 为仿真准备 CAD 模型：使用 CAD-to-SimReady 技能。
  Prepare CAD models for simulation: Use the CAD-to-SimReady skills.
• 验证、打包并提交您的资产：请遵循 SimReady Central 提交流程。
  Validate, package, and submit your assets: Follow the SimReady Central submission skill.

祝你构建愉快，我们迫不及待想看看你创造的作品！Happy building, we can’t wait to see what you create!