A general purpose foundation model for humanoid robot learning.

人形機器人學習的通用基礎模型。

The GR00T model takes multimodal instructions and past interactions as input and produces the next action for the robot to execute.

GR00T模型將多模式指令和過去的交互作為輸入，並生成機器人執行的下一個動作。

We developed Isaac Lab, a robot learning application to train GR00T on Omniverse Isaac Sim.

我們開發了 Isaac Lab，這是一個機器人學習應用程序，用於在 Omniverse Isaac Sim 上訓練 GR00T。

And we scale out with OSMO, a new compute orchestration service that coordinates workflows across DGX systems for training and OVX systems for simulation.

而我們通過 OSMO 進行擴展，這是一項新的計算編排服務，可協調在 DGX 系統上進行訓練和在 OVX 系統上進行模擬的工作流程。

With these tools, we can train GR00T in physically based simulation and transfer zero shot to the real world.

有了這些工具，我們可以在基於物理的模擬中訓練 GR00T，並將零樣本直接應用到現實世界中。

The group model will enable a robot to learn from a handful of human demonstrations so it can help with everyday tasks and emulate human movement just by observing us.

這個群體模型將使機器人能夠從少數人類示範中學習，以便在日常任務中提供幫助，並通過觀察我們來模擬人類的動作。

This is made possible with invidious technologies that can understand humans from videos, train models and simulation and ultimately deploy them directly to physical robots.

這是通過能夠從影片中理解人類、在模擬中訓練模型，最終直接部署到實體機器人上的 NVIDIA 技術所實現的。

Connecting group to a large language model even allows it to generate motions by following natural language instructions.

將群體與大型語言模型連接，甚至可以通過遵循自然語言指令來生成動作。

Hi, GR-1. Can you give me a high-five?

嗨，GR-1，你能跟我擊掌嗎？

Sure thing! Let's high-five.

當然！來擊掌吧。

Can you give us some cool moves?

你能給我們一些很酷的動作嗎？

Sure! Check this out!

當然！你看看！

All this incredible intelligence is powered by the new Jetson Thor robotics chips, designed for GR00T built for the future.

所有這些令人難以置信的智能都由新的Jetson Thor機器人芯片提供動力，這些芯片是專為GR00T設計，面向未來。

With Isaac Lab, OSMO and GR00T, we're providing the building blocks for the next generation of AI-powered robotics.

透過Isaac Lab、OSMO 和 GR00T，我們為下一代 AI 動力機器人提供了構建基礎。

About the same size.

我們差不多大。

The soul of NVIDIA, the intersection of computer graphics, physics, artificial intelligence. It all came to bear at this moment.

NVIDIA的靈魂，計算機圖形、物理、人工智慧的交匯處，在這一刻集中展現出來。

The name of that project, General Robotics 003.

這個項目的名稱是 General Robotics 003。

I know. Super good, super good.

我知道，超級好。

Well, I think we have some special guests, do we?

我想我們有一些特別的客人，是嗎？

Hey guys.

你們好。

So I understand you guys are powered by Jetson.

據我所知，你們是由 Jetson 驅動的。

They're powered by Jetsons.

它們的動力來自 Jetsons。

Little Jetson robotics computers inside.

內有小小的 Jetsons 機器人電腦。

They learn to walk in Isaac Sim.

他們在 Isaac Sim 學會走路。

Ladies and gentlemen, this is Orange and this is the famous Green.

女士們，先生們，這是小橘，這是著名的小綠。

They are the BDX robots of Disney. Amazing Disney research.

他們是迪士尼的BDX機器人。 令人驚嘆的迪士尼研究。

Come on, you guys, let's wrap up, let's go.

來吧，各位，我們收工了，走吧。

Five things.

五件事。

Where are you going? I say right here.

你要去哪兒？我說來這。

Don't be afraid.

別害怕。

Come here, Green.

過來，小綠。

Hurry up.

快點。

What are you saying?

你在說什麼？

No, it's not time to eat.

不，還沒到吃飯時間。

It's not time to eat.

還沒到吃飯時間。

I'll give you, I'll give you a snack in a moment.

我會給你，我馬上給你點心。

Let me finish up real quick.

讓我快點說完。

Come on, Green, hurry up.

綠色，快點。

Stop wasting time.

別浪費時間了。

The next generation of robotics will likely be a humanoid robotics.

下一代機器人技術很可能是仿人機器人技術。

We now have the necessary technology.

我們現在擁有必要的技術。

And as I was describing earlier, the necessary technology to imagine generalized human robotics in a way human robotics is likely easier.

正如我之前所描述的，以人類機器人的方式來想象通用人類機器人所需的技術可能更容易。

And the reason for that is because we have a lot more imitation training data that we can provide the robots because we are constructed in a very similar way.

這是因為我們有更多的模仿訓練數據可以提供給機器人，因為我們的構造非常相似。

It is very likely that the human and robotics will be much more useful in our world because we created the world to be something that we can interoperate in and work well in.

在我們的世界裡，人類和機器人很有可能會發揮更大的作用，因為我們創造的世界是可以讓我們在其中互通有無、合作無間的。

And the way that we set up our workstations and manufacturing and logistics they were designed for, for humans, they were designed for people.

我們設置工作站、生產和物流的方式是為人類設計的，是為人設計的，

And so these human robotics will likely be much more productive to deploy.

所以這些人類機器人的部署可能會更有成效。

While we're creating, just like we're doing with the others, the entire stack starting from the top, a foundation model that learns from watching video, human examples.

在我們創造時，就像我們與其他人一樣，從頂部開始的整個堆棧，一個基礎模型，從觀看影片、人類示例中學習。

It could be in video form, it could be in virtual reality form.

它可以是影片形式，也可以是虛擬現實形式。

We then created a gym for it called Isaac Reinforcement Learning Gym, which allows the humanoid robot to learn how to adapt to the physical world.

然後我們為其創建了一個名為Isaac Reinforcement Learning Gym的場館，這使得人形機器人能夠學習如何適應物理世界。

And then, an incredible computer—the same computer that's going to go into a robotic car—will run inside a humanoid robot called Thor.

然後，一台不可思議的電腦——將與機器人汽車相同的電腦——將運用在一個名為Thor的人形機器人內部。

It's designed for transformer engines.

這是為變壓器引擎而設計的。