Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Training a Whole-Body Control Foundation Model -- new work from my team at Agility A neural network for controlling our humanoid robots which is robust to disturbances, can handle heavy objects, and is a powerful platform for learning new whole-body skills learn more in our blog post ->

Chris Paxton

37,093 subscribers

46,193 views • 1 year ago •via X (Twitter)

Science & Technology Gaming Arts

Anya Rossi• Live Now

Private livecam show

22 Comments

Chris Paxton1 year ago

Blog post on whole-body control:

Chris Paxton1 year ago

Yes those are cinderblocks! One of the advantages of this control strategy is that it's pretty simple to adapt to heavy objects, like we'd want to do for robots to be useful in the kinds of difficult jobs we want to automate.

kache1 year ago

@agilityrobotics this is insanely difficult

Chris Paxton1 year ago

@agilityrobotics Yeah it's been a long road! I'm very excited about where it's ended up though! All credit to @JonahSiekmann here

Maxime Chevalier1 year ago

@agilityrobotics At first I thought it was CGI when I saw the thumbnail because that white room looks so clean/plain! Glad to see you guys are adding hands to your robots! :)

Gabriel Diaz 🇧🇷1 year ago

@agilityrobotics $PEAQ 🔥🔥🔥

Tom Ciszek1 year ago

@agilityrobotics Motor cortex for machines that can lift, carry, adapt ... AND stay standing. Congrats team. Huge milestone. 🦾

R2rule1 year ago

@agilityrobotics It's amazing how your CEO can find investors for those stupid legs and make you run to the top with those inefficient legs. Looks like nobody has balls to say the truth about that stupid and inefficient legs to your boss.

Bercan1 year ago

@agilityrobotics I love this! Real2sim , sim2real has soo much potential. Imagine doing this for HVAC technicians or plumbers

Uking Sun1 year ago

@agilityrobotics The hand design is interesting.

Kelechi RecordBreaker1 year ago

@agilityrobotics Whole-body control foundation models are crucial for adaptable humanoid robotics. This work could set new benchmarks for real-world deployment. Well done!

PrismaX1 year ago

@agilityrobotics This one is genuinely strong. GG.

Emerson Segura1 year ago

@agilityrobotics nice work

Phil Trubey1 year ago

@agilityrobotics Oh wow. When did you join them?

Chris Paxton1 year ago

@agilityrobotics Start of year

Barry Sayer 💻🖱️☁️🤖ᯅ 🐙1 year ago

@agilityrobotics Amazing work. Been waiting on cart pushing to make a comeback for years.

CIX 🦾1 year ago

@agilityrobotics Awesome work

Chris Paxton1 year ago

@agilityrobotics Thanks!

Quentin Munch1 year ago

@agilityrobotics Very impressive work, I can understand how position teaching reward can be generated for the arm and torso, but for the leg ? Is it really control from a position in robot workspace (no velocity) ?

Himanshu Kumar1 year ago

@agilityrobotics Interesting approach. A more generalized control model could unlock so much potential in robotics.

HDP1 year ago

@agilityrobotics yo bro look like he shit his pants fr

arun kumar singh1 year ago

@agilityrobotics Is everything a foundational model now? There are now task specific e2e controllers?

Related Videos

🤖Introducing TWIST: Teleoperated Whole-Body Imitation System. We develop a humanoid teleoperation system to enable coordinated, versatile, whole-body movements, using a single neural network. This is our first step toward general-purpose robots. 🌐

🤖Introducing TWIST: Teleoperated Whole-Body Imitation System. We develop a humanoid teleoperation system to enable coordinated, versatile, whole-body movements, using a single neural network. This is our first step toward general-purpose robots. 🌐

Yanjie Ze

80,217 views • 1 year ago

Chinese robotics company Unitree has developed a new AI model for general-purpose humanoid robots. The robot can autonomously sort objects, arrange flowers, stack cups, handle laundry, and organize shoes. The system combines vision, reasoning, and whole-body control to perform a wide range of household tasks.

Chinese robotics company Unitree has developed a new AI model for general-purpose humanoid robots. The robot can autonomously sort objects, arrange flowers, stack cups, handle laundry, and organize shoes. The system combines vision, reasoning, and whole-body control to perform a wide range of household tasks.

Space and Technology

31,172 views • 9 days ago

Robots need to be able to apply pressure and make contact with objects as needed in order to accomplish their tasks. From compliance to working safely around humans to whole-body manipulation of heavy objects, combining force and position control can dramatically expand the capabilities of robots. This is especially true for legged robots, which have so much ability to exert forces on the world around them. But how do we train robots which can do this? Baoxiong Jia tells us more in our discussion of his team’s recent, Best Paper Award winning work on learning a unified policy for position and force control, called UniFP. To learn more, watch Episode #49 of RoboPapers, hosted by Michael Cho - Rbt/Acc and Chris Paxton.

Robots need to be able to apply pressure and make contact with objects as needed in order to accomplish their tasks. From compliance to working safely around humans to whole-body manipulation of heavy objects, combining force and position control can dramatically expand the capabilities of robots. This is especially true for legged robots, which have so much ability to exert forces on the world around them. But how do we train robots which can do this? Baoxiong Jia tells us more in our discussion of his team’s recent, Best Paper Award winning work on learning a unified policy for position and force control, called UniFP. To learn more, watch Episode #49 of RoboPapers, hosted by Michael Cho - Rbt/Acc and Chris Paxton.

RoboPapers

44,825 views • 9 months ago

🎾Your humanoid tennis player is here!🤖 Introducing LATENT (Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data) — the world’s first real-time whole-body planning and control algorithm for athletic humanoid tennis. For the first time, a humanoid robot can sustain high-dynamic, long-horizon tennis rallies with millisecond-level reactions, precise ball striking, and natural whole-body motion. This marks a leap from mechanical motion imitation to intelligent, decision-driven athletic interaction. #Galbot #Robotics #Innovation #AI #Technology #Humanoid

🎾Your humanoid tennis player is here!🤖 Introducing LATENT (Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data) — the world’s first real-time whole-body planning and control algorithm for athletic humanoid tennis. For the first time, a humanoid robot can sustain high-dynamic, long-horizon tennis rallies with millisecond-level reactions, precise ball striking, and natural whole-body motion. This marks a leap from mechanical motion imitation to intelligent, decision-driven athletic interaction. #Galbot #Robotics #Innovation #AI #Technology #Humanoid

Galbot

46,123 views • 6 months ago

How do you give a humanoid the general motion capability? Not just single motions, but all motion? Introducing SONIC, our new work on supersizing motion tracking for natural humanoid control. We argue that motion tracking is the scalable foundation task for humanoids. So we "supersized" it: 9k+ GPU hours and 100M+ motion frames. But tracking alone is not enough; we show how to make a useful control system out of it: - Universal Kinematic Planner: Enables game-like gamepad control and high-level teleoperation, just like controlling a character in a game. - VR Full-Body Teleop: Direct, real-time whole-body control by a human wearing a VR headset. - VR Keypoint Teleop: Control the upper body (hands/head) while our planner handles robust locomotion automatically. - VLA Integration: We connect this motion tracker to autonomous Visual-Language-Action (VLA) models for autonomous task execution! We use a Universal Token Space to UNIFY this command space, turning our robust tracker into a general-purpose, programmable humanoid brain. This is the generalist "System 1" for humanoids. 🚀 Project: #Humanoids #Robotics #AI #FoundationModels #NVIDIAResearch 🧠🔥

How do you give a humanoid the general motion capability? Not just single motions, but all motion? Introducing SONIC, our new work on supersizing motion tracking for natural humanoid control. We argue that motion tracking is the scalable foundation task for humanoids. So we "supersized" it: 9k+ GPU hours and 100M+ motion frames. But tracking alone is not enough; we show how to make a useful control system out of it: - Universal Kinematic Planner: Enables game-like gamepad control and high-level teleoperation, just like controlling a character in a game. - VR Full-Body Teleop: Direct, real-time whole-body control by a human wearing a VR headset. - VR Keypoint Teleop: Control the upper body (hands/head) while our planner handles robust locomotion automatically. - VLA Integration: We connect this motion tracker to autonomous Visual-Language-Action (VLA) models for autonomous task execution! We use a Universal Token Space to UNIFY this command space, turning our robust tracker into a general-purpose, programmable humanoid brain. This is the generalist "System 1" for humanoids. 🚀 Project: #Humanoids #Robotics #AI #FoundationModels #NVIDIAResearch 🧠🔥

Zhengyi “Zen” Luo

62,703 views • 10 months ago

Congratulations to our NVIDIA Research team for SONIC being published in Science Magazine. 🦾 Built by NVIDIA Robotics researchers and collaborators, SONIC demonstrates scaling motion tracking for natural, robust humanoid whole-body control. Read the paper 📄

Congratulations to our NVIDIA Research team for SONIC being published in Science Magazine. 🦾 Built by NVIDIA Robotics researchers and collaborators, SONIC demonstrates scaling motion tracking for natural, robust humanoid whole-body control. Read the paper 📄

NVIDIA Robotics

52,565 views • 1 month ago

THE CRUCIBLE IS NOW LIVE Download - The Seneschal awaits! Experience a whole new Cerim Crucible & so much more! Check our blog post for more -

THE CRUCIBLE IS NOW LIVE Download - The Seneschal awaits! Experience a whole new Cerim Crucible & so much more! Check our blog post for more -

No Rest for the Wicked

155,180 views • 2 years ago

🤖✨Excited to share our new work: OMG: Omni-Modal Motion Generation for Generalist Humanoid Control What if a humanoid could understand intent from language, music/audio, human motion, or their combinations—and turn it into executable whole-body motion in real time? [🧵1/11]

🤖✨Excited to share our new work: OMG: Omni-Modal Motion Generation for Generalist Humanoid Control What if a humanoid could understand intent from language, music/audio, human motion, or their combinations—and turn it into executable whole-body motion in real time? [🧵1/11]

Siqiao Huang

13,750 views • 3 months ago

1x Neo Gamma getting a bottle of water from the fridge. Being able to open something like a heavy stainless steel fridge door is a quiet advantage of a humanoid with whole-body control; humans use our whole bodies to accomplish even very simple tasks like this.

1x Neo Gamma getting a bottle of water from the fridge. Being able to open something like a heavy stainless steel fridge door is a quiet advantage of a humanoid with whole-body control; humans use our whole bodies to accomplish even very simple tasks like this.

Chris Paxton

34,246 views • 1 year ago

Researchers have developed Humanoid-GPT, a new AI system that helps humanoid robots perform complex movements and tasks. Tested on the Unitree G1 robot, it enables real-time whole-body control and can perform new actions without special training. During demonstrations, the robot successfully carried out a variety of movements and tasks, showing strong potential for real-world applications.

Researchers have developed Humanoid-GPT, a new AI system that helps humanoid robots perform complex movements and tasks. Tested on the Unitree G1 robot, it enables real-time whole-body control and can perform new actions without special training. During demonstrations, the robot successfully carried out a variety of movements and tasks, showing strong potential for real-world applications.

Space and Technology

22,330 views • 3 months ago

🚀 Can we make a humanoid move like Cristiano Ronaldo, LeBron James and Kobe Byrant? YES! 🤖 Introducing ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills Website: Code:

🚀 Can we make a humanoid move like Cristiano Ronaldo, LeBron James and Kobe Byrant? YES! 🤖 Introducing ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills Website: Code:

Tairan He

259,205 views • 1 year ago

Reinforcement learning has unlocked tremendous gains in broader AI and humanoid robot whole body control, but still has yet to deliver strong results for robot manipulation due to the difficulty of capturing tasks in an accurate simulation; one way to fix this is to create complex environments with learned world models. But learning fully within world models is often intractable as powerful world models are computationally too expensive. Instead, Joseph Amigo and Rooholla Khorrambakht propose to break the problem down: into a large-scale world model that can generate forward trajectories, and a lightweight, low-dimensional latent-space model which can approximate local dynamics of a problem, without needing to back-propagate through a heavy global model. What this means: you can do reinforcement learning in this coupled local-global world model, and learn contact rich skills for a robot. To learn more, watch Episode #101 of RoboPapers, with Michael Cho - Rbt/Acc and Jiafei Duan!

Reinforcement learning has unlocked tremendous gains in broader AI and humanoid robot whole body control, but still has yet to deliver strong results for robot manipulation due to the difficulty of capturing tasks in an accurate simulation; one way to fix this is to create complex environments with learned world models. But learning fully within world models is often intractable as powerful world models are computationally too expensive. Instead, Joseph Amigo and Rooholla Khorrambakht propose to break the problem down: into a large-scale world model that can generate forward trajectories, and a lightweight, low-dimensional latent-space model which can approximate local dynamics of a problem, without needing to back-propagate through a heavy global model. What this means: you can do reinforcement learning in this coupled local-global world model, and learn contact rich skills for a robot. To learn more, watch Episode #101 of RoboPapers, with Michael Cho - Rbt/Acc and Jiafei Duan!

RoboPapers

16,760 views • 22 days ago

SkildAI, which is developing a foundational AI model for robots, has released a new video of their humanoid robot walking over objects and exploring environments it had never seen before. The robot is using vision & adapts in real time. SkildAI: "From raw images and joint feedback, the model directly outputs low-level motor commands. This single neural network enables humanoid robots to seamlessly walk across flat ground, climb stairs, and step over obstacles without any planning, mapping, or manual switching between behaviors. For all of our testing in this video, we deployed the robot in each new environment without any prior planning or mapping. We took off the humanoid’s shoes so you can hear it.

SkildAI, which is developing a foundational AI model for robots, has released a new video of their humanoid robot walking over objects and exploring environments it had never seen before. The robot is using vision & adapts in real time. SkildAI: "From raw images and joint feedback, the model directly outputs low-level motor commands. This single neural network enables humanoid robots to seamlessly walk across flat ground, climb stairs, and step over obstacles without any planning, mapping, or manual switching between behaviors. For all of our testing in this video, we deployed the robot in each new environment without any prior planning or mapping. We took off the humanoid’s shoes so you can hear it.

Sawyer Merritt

69,381 views • 1 year ago

We’ve raised $50 million in Series A funding from DST Global Partners, NVIDIA (NVIDIA’s venture capital arm), redalpine , prosusventures , and Moonfire Ventures , following our $7.35M seed round from Frst , Moonfire Ventures , and redalpine just a few months earlier, to build the autonomy stack that makes humanoid robots adaptive, intelligent, and ready for real-world deployment at scale. In less than a year, our team has shown that long-horizon whole-body humanoid control can scale across hardware and tasks by leveraging the power of simulation and reinforcement learning. This funding will help us grow our team, scale our compute and robot fleets, and accelerate the commercialization of our autonomy stack with OEM partners globally. You can find more details in the links shared in the comments.

We’ve raised $50 million in Series A funding from DST Global Partners, NVIDIA (NVIDIA’s venture capital arm), redalpine , prosusventures , and Moonfire Ventures , following our $7.35M seed round from Frst , Moonfire Ventures , and redalpine just a few months earlier, to build the autonomy stack that makes humanoid robots adaptive, intelligent, and ready for real-world deployment at scale. In less than a year, our team has shown that long-horizon whole-body humanoid control can scale across hardware and tasks by leveraging the power of simulation and reinforcement learning. This funding will help us grow our team, scale our compute and robot fleets, and accelerate the commercialization of our autonomy stack with OEM partners globally. You can find more details in the links shared in the comments.

Flexion

73,827 views • 10 months ago

TWIST: a real-time teleoperation system for humanoid robots to mimic whole-body motions. Reference motion data is generated by retargeting human motion-capture data to the robot. Then, the controller is trained in simulation using reinforcement learning and behavior cloning.

TWIST: a real-time teleoperation system for humanoid robots to mimic whole-body motions. Reference motion data is generated by retargeting human motion-capture data to the robot. Then, the controller is trained in simulation using reinforcement learning and behavior cloning.

The Humanoid Hub

46,715 views • 1 year ago

How can we build a general-purpose “foundation model” for robot motion? Zhengyi “Zen” Luo joins us to talk about SONIC, which uses motion tracking as a foundational task for humanoid robot control, and scales humanoid control training to 9k GPU hours and 100 million frames worth of data. The result: a model with a generally-useful embedding space that can be controlled by a VLA, or from human video, to perform a wide variety of humanoid whole-body-control tasks, including with zero-shot transfer to previously unseen motions. Watch episode 72 of RoboPapers, with Michael Cho - Rbt/Acc and Jiafei Duan, now!

How can we build a general-purpose “foundation model” for robot motion? Zhengyi “Zen” Luo joins us to talk about SONIC, which uses motion tracking as a foundational task for humanoid robot control, and scales humanoid control training to 9k GPU hours and 100 million frames worth of data. The result: a model with a generally-useful embedding space that can be controlled by a VLA, or from human video, to perform a wide variety of humanoid whole-body-control tasks, including with zero-shot transfer to previously unseen motions. Watch episode 72 of RoboPapers, with Michael Cho - Rbt/Acc and Jiafei Duan, now!

RoboPapers

24,802 views • 5 months ago