Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

Ego-VCP: a learned ego-vision world model trained offline on demonstration-free random data that predicts dynamics in latent space for humanoid robots. Achieved robust real-time contact-rich planning on a real Unitree G1: bracing against walls, blocking flying objects, traversing low arches. Paper:

The Humanoid Hub

83,193 subscribers

24,742 görüntüleme • 9 ay önce •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

A new paper by Disney Research introduces a 2-stage technique for motion tracking in physics-based character animation, using a pretrained latent space and RL to control diverse and unseen motions on both virtual and real-world humanoid robots. Paper:

A new paper by Disney Research introduces a 2-stage technique for motion tracking in physics-based character animation, using a pretrained latent space and RL to control diverse and unseen motions on both virtual and real-world humanoid robots. Paper:

The Humanoid Hub

20,088 görüntüleme • 2 yıl önce

Researchers developed Hitter, a humanoid robot that can play table tennis in real-world settings. It was trained on the Unitree G1 robot to predict the ball’s path, choose the right shot, and react in less than a second. In tests, it achieved 106 consecutive shots with a human and even rallied with another humanoid robot.

Researchers developed Hitter, a humanoid robot that can play table tennis in real-world settings. It was trained on the Unitree G1 robot to predict the ball’s path, choose the right shot, and react in less than a second. In tests, it achieved 106 consecutive shots with a human and even rallied with another humanoid robot.

Space and Technology

25,736 görüntüleme • 4 ay önce

Unitree Strikes Double Gold on Day One🥇🥇🏃 The First World Humanoid Robot Games Unitree takes first place in all Day-One races 1500m track race — 6:34.40 (Unweighted) (Unitree H1 humanoid robot — the same model featured in the Spring Festival Gala) 400m track race — 1:28.03 (Unweighted real stopwatch time 1:23.03) Let’s set a new running speed record for humanoid robots together!

Unitree Strikes Double Gold on Day One🥇🥇🏃 The First World Humanoid Robot Games Unitree takes first place in all Day-One races 1500m track race — 6:34.40 (Unweighted) (Unitree H1 humanoid robot — the same model featured in the Spring Festival Gala) 400m track race — 1:28.03 (Unweighted real stopwatch time 1:23.03) Let’s set a new running speed record for humanoid robots together!

Unitree

13,799,445 görüntüleme • 1 yıl önce

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

22,234 görüntüleme • 2 ay önce

In just ~3 months, as a solo founder with no prior robotics experience, General Trajectory trained a foundation model for dexterous manipulation that lets humanoid robots pick up unseen objects and perform real-world work. It generalizes to novel objects and scenes, including cases where prior SoTA models achieve 0% success. Congrats on the launch Joshua!

In just ~3 months, as a solo founder with no prior robotics experience, General Trajectory trained a foundation model for dexterous manipulation that lets humanoid robots pick up unseen objects and perform real-world work. It generalizes to novel objects and scenes, including cases where prior SoTA models achieve 0% success. Congrats on the launch Joshua!

Y Combinator

68,639 görüntüleme • 7 ay önce

Real-world robot data is expensive and slow to collect, creating a major challenge for humanoid development. 🤖 The NVIDIA GR00T N1.6 open vision language action model is pre-trained on a diverse mix of data, including thousands of hours of Stanford Vision and Learning Lab’s BEHAVIOR simulation data, which covers long-horizon everyday manipulation tasks. This diverse training is the key to robust cross-embodiment performance and real-world adaptability. 🌍 Read the blog 🔗

Real-world robot data is expensive and slow to collect, creating a major challenge for humanoid development. 🤖 The NVIDIA GR00T N1.6 open vision language action model is pre-trained on a diverse mix of data, including thousands of hours of Stanford Vision and Learning Lab’s BEHAVIOR simulation data, which covers long-horizon everyday manipulation tasks. This diverse training is the key to robust cross-embodiment performance and real-world adaptability. 🌍 Read the blog 🔗

NVIDIA Robotics

13,479 görüntüleme • 7 ay önce

A robot that sees the terrain and predicts its own future… up to 5 seconds ahead? This is real. ❗️Best Systems Paper finalist at #RSS2025 The team introduces a perceptive Forward Dynamics Model that helps legged robots safely navigate rough, complex environments: no manual tuning no rigid-body limits. Why this matters ✅ Predicts future robot states using both perception and proprioception ✅ Trained on years of sim + real data for strong sim-to-real performance ✅ Enables zero-shot navigation with minimal cost tuning ✅ Boosts success rates and safety on rough terrain like never before This isn’t just another planner, it’s a big step toward robust, adaptable locomotion. Paper: Project website:

A robot that sees the terrain and predicts its own future… up to 5 seconds ahead? This is real. ❗️Best Systems Paper finalist at #RSS2025 The team introduces a perceptive Forward Dynamics Model that helps legged robots safely navigate rough, complex environments: no manual tuning no rigid-body limits. Why this matters ✅ Predicts future robot states using both perception and proprioception ✅ Trained on years of sim + real data for strong sim-to-real performance ✅ Enables zero-shot navigation with minimal cost tuning ✅ Boosts success rates and safety on rough terrain like never before This isn’t just another planner, it’s a big step toward robust, adaptable locomotion. Paper: Project website:

Ilir Aliu

43,990 görüntüleme • 1 yıl önce

🤖ROBOTS ARE GETTING SMARTER AT TOUCHING THE REAL WORLD Researchers from UC Berkeley, NVIDIA and Stanford introduced T-Rex, a framework that combines vision, language and tactile sensing. Instead of relying on cameras alone, robots can now respond to physical contact in real time. Robots are no longer just seeing objects. They’re learning how to FEEL them.

🤖ROBOTS ARE GETTING SMARTER AT TOUCHING THE REAL WORLD Researchers from UC Berkeley, NVIDIA and Stanford introduced T-Rex, a framework that combines vision, language and tactile sensing. Instead of relying on cameras alone, robots can now respond to physical contact in real time. Robots are no longer just seeing objects. They’re learning how to FEEL them.

Coin Bureau

28,275 görüntüleme • 2 ay önce

Meet Any2Track, a new two-stage RL framework designed to address a core challenge for humanoid robots: reliably tracking diverse, dynamic motions in the real world. The framework consists of two key components: AnyTracker, a general motion tracker, and AnyAdapter, a module that allows the robot to adapt to real-world disturbances like uneven terrain or external forces. This approach gives robots the stability needed for practical use. The system has been successfully deployed on a Unitree G1 humanoid robot with zero-shot sim2real transfer, performing exceptionally well in various motion tracking tasks under multiple real-world disturbances. Project: Codebase:

Meet Any2Track, a new two-stage RL framework designed to address a core challenge for humanoid robots: reliably tracking diverse, dynamic motions in the real world. The framework consists of two key components: AnyTracker, a general motion tracker, and AnyAdapter, a module that allows the robot to adapt to real-world disturbances like uneven terrain or external forces. This approach gives robots the stability needed for practical use. The system has been successfully deployed on a Unitree G1 humanoid robot with zero-shot sim2real transfer, performing exceptionally well in various motion tracking tasks under multiple real-world disturbances. Project: Codebase:

RoboHub🤖

61,378 görüntüleme • 11 ay önce

Hot take: robots should not dream in pixels. Pixels are too low-level. Latents are too opaque. μ₀ predicts a third thing: 3D motion traces. On real robots, it beats π₀.₅ — with ~1/100 the data scale and no action labels for world-model pretraining. 🧵 (this video features voiceover narration)

Hot take: robots should not dream in pixels. Pixels are too low-level. Latents are too opaque. μ₀ predicts a third thing: 3D motion traces. On real robots, it beats π₀.₅ — with ~1/100 the data scale and no action labels for world-model pretraining. 🧵 (this video features voiceover narration)

Furong Huang

113,285 görüntüleme • 2 ay önce

Chinese robotics company Astribot released their latest World-Action Model (WAM), Lumo-2. Technical breakdown: - based on a frozen 🥶 Qwen-3.5 4B VLM - trained in 3 progressive stages: 1. Action is aligned with latent world dynamics (an abstract representation of action). Real-world actions are anchored to physical constraints, while the latent space is guided to focus on motion-relevant changes. This bidirectional relationship makes the model physically grounded -> critical for a world model. 2. Action is aligned with vision and language. Reusing the vision backbone and action encoder from the frozen VLM, the authors add a custom vocabulary (for new actions), a semantic module, an action decoder, and an action projector. This aligns the (new) action representations with the (existing) vision-language semantic space. Most importantly: it builds a direct mapping from natural-language instructions to motor execution. 3. End-to-end training on language, video, and robot data. Only the new modules (everything outside the frozen backbone) are trained end-to-end across temporal reasoning, physical understanding, long-horizon, and dexterous manipulation. At the end of the day, Lumo-2 is not the best on benchmarks, but that's not the point. What's genuinely new: - a way to combine latent world modeling and action generation through progressive alignment - a physically-grounded latent dynamics space - it lifts performance on unseen objects using un-annotated human egocentric video + Vision Pro captures, no special transfer algorithm needed Why it matters: - the whole model is thin trainable adapters (semantic module, action decoder/projector) on a frozen 4B backbone (cheap) - that scale is suited for real-time embedded inference (~2.71× decode speedup, no accuracy loss) - its real moat is long-horizon execution, where the added temporal memory pays off far more than on any other task As a result, this robot can now make your latte (5x sped up video):

Chinese robotics company Astribot released their latest World-Action Model (WAM), Lumo-2. Technical breakdown: - based on a frozen 🥶 Qwen-3.5 4B VLM - trained in 3 progressive stages: 1. Action is aligned with latent world dynamics (an abstract representation of action). Real-world actions are anchored to physical constraints, while the latent space is guided to focus on motion-relevant changes. This bidirectional relationship makes the model physically grounded -> critical for a world model. 2. Action is aligned with vision and language. Reusing the vision backbone and action encoder from the frozen VLM, the authors add a custom vocabulary (for new actions), a semantic module, an action decoder, and an action projector. This aligns the (new) action representations with the (existing) vision-language semantic space. Most importantly: it builds a direct mapping from natural-language instructions to motor execution. 3. End-to-end training on language, video, and robot data. Only the new modules (everything outside the frozen backbone) are trained end-to-end across temporal reasoning, physical understanding, long-horizon, and dexterous manipulation. At the end of the day, Lumo-2 is not the best on benchmarks, but that's not the point. What's genuinely new: - a way to combine latent world modeling and action generation through progressive alignment - a physically-grounded latent dynamics space - it lifts performance on unseen objects using un-annotated human egocentric video + Vision Pro captures, no special transfer algorithm needed Why it matters: - the whole model is thin trainable adapters (semantic module, action decoder/projector) on a frozen 4B backbone (cheap) - that scale is suited for real-time embedded inference (~2.71× decode speedup, no accuracy loss) - its real moat is long-horizon execution, where the added temporal memory pays off far more than on any other task As a result, this robot can now make your latte (5x sped up video):

Léo

32,296 görüntüleme • 1 ay önce

The power of generative models — now embodied in humanoids. Announcing DreamControl –– After a year-long research effort at General Robotics — we present a scalable framework for whole-body humanoid control that fuses diffusion priors with reinforcement learning to unlock real-world scene interaction. Diffusion + RL → natural whole-body skills on real robots. DreamControl enables humanoids to move beyond locomotion demos → performing natural, human-like skills such as –– Picking & lifting objects, Opening drawers & doors, Precise punching, kicking, and jumping, Bimanual manipulation tasks Our key innovation: a diffusion prior over human motion that guides RL, eliminating the need for massive teleoperation datasets, and producing motions that look human while transferring to real hardware. Trained purely in simulation, deployed on the Unitree G1 humanoid, DreamControl policies run in real time, bridging sim-to-real with unprecedented naturalness. We leverage a novel hybrid edge + cloud infrastructure that runs RL-trained policies on the edge backed by powerful AI models running in the cloud This is the next step in General Robotics’ journey toward general-purpose humanoid assistants that interact, adapt, and assist autonomously. Paper: Blog: 1/n

The power of generative models — now embodied in humanoids. Announcing DreamControl –– After a year-long research effort at General Robotics — we present a scalable framework for whole-body humanoid control that fuses diffusion priors with reinforcement learning to unlock real-world scene interaction. Diffusion + RL → natural whole-body skills on real robots. DreamControl enables humanoids to move beyond locomotion demos → performing natural, human-like skills such as –– Picking & lifting objects, Opening drawers & doors, Precise punching, kicking, and jumping, Bimanual manipulation tasks Our key innovation: a diffusion prior over human motion that guides RL, eliminating the need for massive teleoperation datasets, and producing motions that look human while transferring to real hardware. Trained purely in simulation, deployed on the Unitree G1 humanoid, DreamControl policies run in real time, bridging sim-to-real with unprecedented naturalness. We leverage a novel hybrid edge + cloud infrastructure that runs RL-trained policies on the edge backed by powerful AI models running in the cloud This is the next step in General Robotics’ journey toward general-purpose humanoid assistants that interact, adapt, and assist autonomously. Paper: Blog: 1/n

Ashish Kapoor

118,357 görüntüleme • 11 ay önce

DiT4DiT is now open source! As the first humanoid-deployable Video-Action Model built on a world model, DiT4DiT continues to surprise us. In our paper last month, we showed its strong data efficiency. Now, with only slight modifications, it enables real-time whole-body autonomous pick-and-place. Paper: Code: Website:

DiT4DiT is now open source! As the first humanoid-deployable Video-Action Model built on a world model, DiT4DiT continues to surprise us. In our paper last month, we showed its strong data efficiency. Now, with only slight modifications, it enables real-time whole-body autonomous pick-and-place. Paper: Code: Website:

Shuo Yang

14,534 görüntüleme • 4 ay önce

DexNDM — a neural dynamics model for dexterous hands Galbot, Tsinghua University and Shanghai Qizhi Institute released DexNDM, a neuromuscular-style dynamics model for dexterous hands. The team reports it reduces reliance on large sets of flawless demonstrations: using biased real-world data, DexNDM bridges Sim→Real for in-hand rotations across arbitrary poses and axes, handling slender, tiny, and irregular objects. Developers say the method is moving from lab validation toward production-ready workflows.

DexNDM — a neural dynamics model for dexterous hands Galbot, Tsinghua University and Shanghai Qizhi Institute released DexNDM, a neuromuscular-style dynamics model for dexterous hands. The team reports it reduces reliance on large sets of flawless demonstrations: using biased real-world data, DexNDM bridges Sim→Real for in-hand rotations across arbitrary poses and axes, handling slender, tiny, and irregular objects. Developers say the method is moving from lab validation toward production-ready workflows.

RoboHub🤖

34,445 görüntüleme • 9 ay önce

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 görüntüleme • 1 yıl önce

Turning video into humanoid robot motion! 🤳🏼 Training humanoid robots needs huge amounts of motion data, but real-world capture doesn’t scale. Mocap is expensive, dangerous edge cases are rare, and you can’t ask humans to repeatedly fall or crash. Video2Robot tackles this by converting videos into physics-grounded humanoid simulations. Motion is generated to respect balance, inertia, ground contact, and joint limits, then directly retargeted to robot simulators. One prompt can generate a full humanoid motion sequence, including multi-agent interactions and failure cases like falls or collisions, scenarios that are hard or impossible to capture safely in the real world. The pipeline is model-agnostic and works with existing video generators, making it a practical way to scale data for robots. If robots are going to operate in the real world, they need to be trained on the failures too, not just the perfect demos. Here's the GitHub: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Turning video into humanoid robot motion! 🤳🏼 Training humanoid robots needs huge amounts of motion data, but real-world capture doesn’t scale. Mocap is expensive, dangerous edge cases are rare, and you can’t ask humans to repeatedly fall or crash. Video2Robot tackles this by converting videos into physics-grounded humanoid simulations. Motion is generated to respect balance, inertia, ground contact, and joint limits, then directly retargeted to robot simulators. One prompt can generate a full humanoid motion sequence, including multi-agent interactions and failure cases like falls or collisions, scenarios that are hard or impossible to capture safely in the real world. The pipeline is model-agnostic and works with existing video generators, making it a practical way to scale data for robots. If robots are going to operate in the real world, they need to be trained on the failures too, not just the perfect demos. Here's the GitHub: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

59,835 görüntüleme • 8 ay önce

Perhaps the "GPT-3 moment" for humanoid robots lies within vast amounts of human video data. UC Berkeley (Abbeel + Malik group) just released Do as I Do: a pipeline that takes monocular RGB human videos (ego + exo)->reconstructs hand-object interactions->retargets them to multi-fingered robotic hands, generating executable robot trajectories. It pushes success rates on real-world clips from ~25% to 71%, outperforming prior SOTA. This kind of human video->robot-complete data approach feels increasingly important for closing the data flywheel in humanoid/embodied AI.

Perhaps the "GPT-3 moment" for humanoid robots lies within vast amounts of human video data. UC Berkeley (Abbeel + Malik group) just released Do as I Do: a pipeline that takes monocular RGB human videos (ego + exo)->reconstructs hand-object interactions->retargets them to multi-fingered robotic hands, generating executable robot trajectories. It pushes success rates on real-world clips from ~25% to 71%, outperforming prior SOTA. This kind of human video->robot-complete data approach feels increasingly important for closing the data flywheel in humanoid/embodied AI.

CyberRobo

12,107 görüntüleme • 2 ay önce

NEWS: DeepMind teaches robots to play soccer. In a paper released today Google's DeepMind details how they used Deep Reinforcement Learning (Deep RL) to train low-cost, miniature humanoid robots in dynamic environments, allowing them to play a simplified one-versus-one (1v1) soccer game. The robots, equipped with 20 actuated joints, were initially trained in simulation using the MuJoCo physics engine. Through this training, they learned robust and dynamic movement skills like rapid fall recovery, walking, turning, and kicking. The robots seamlessly transitioned between these skills, even surpassing expectations, and developed a basic strategic understanding of the game. During matches, the trained robots demonstrated agile skills such as turning, kicking moving balls, and dynamic defensive blocking. They quickly combined these skills, showcasing their adaptability and outperforming scripted baselines. The robots walked faster, got up quicker, and kicked faster than their counterparts. Individual skills were first trained in isolation within the simulation environment, and then composed in a self-play setting. The robots successfully transferred these skills to real-world scenarios. Full paper can be found here:

NEWS: DeepMind teaches robots to play soccer. In a paper released today Google's DeepMind details how they used Deep Reinforcement Learning (Deep RL) to train low-cost, miniature humanoid robots in dynamic environments, allowing them to play a simplified one-versus-one (1v1) soccer game. The robots, equipped with 20 actuated joints, were initially trained in simulation using the MuJoCo physics engine. Through this training, they learned robust and dynamic movement skills like rapid fall recovery, walking, turning, and kicking. The robots seamlessly transitioned between these skills, even surpassing expectations, and developed a basic strategic understanding of the game. During matches, the trained robots demonstrated agile skills such as turning, kicking moving balls, and dynamic defensive blocking. They quickly combined these skills, showcasing their adaptability and outperforming scripted baselines. The robots walked faster, got up quicker, and kicked faster than their counterparts. Individual skills were first trained in isolation within the simulation environment, and then composed in a self-play setting. The robots successfully transferred these skills to real-world scenarios. Full paper can be found here:

Dave Lee

372,039 görüntüleme • 3 yıl önce

Humanoid robots can now play tennis in the style of Nadal, Federer, and Djokovic, learned from broadcast footage. AdaPT (Noitom Robotics, Shanghai AI Lab, Dobot, SJTU) extracts professional player motions from TV broadcasts, retargets them to a humanoid, and learns both rallying and serving while preserving each player's distinctive style. The core problem: decoupled planning-and-tracking works in sim but degrades badly on real hardware. Their fix is speed adaptation, training the tracker on randomized execution speeds while the planner learns to adjust motion speed to the incoming ball. Deployed on Unitree G1 and the full-size Dobot Atom. Worth noting on the sensing: the robot has no onboard vision. Ball and robot position both come from external infrastructure, a 35-camera motion-capture arena in the lab. Paper:

Humanoid robots can now play tennis in the style of Nadal, Federer, and Djokovic, learned from broadcast footage. AdaPT (Noitom Robotics, Shanghai AI Lab, Dobot, SJTU) extracts professional player motions from TV broadcasts, retargets them to a humanoid, and learns both rallying and serving while preserving each player's distinctive style. The core problem: decoupled planning-and-tracking works in sim but degrades badly on real hardware. Their fix is speed adaptation, training the tracker on randomized execution speeds while the planner learns to adjust motion speed to the incoming ball. Deployed on Unitree G1 and the full-size Dobot Atom. Worth noting on the sensing: the robot has no onboard vision. Ball and robot position both come from external infrastructure, a 35-camera motion-capture arena in the lab. Paper:

The Humanoid Hub

85,497 görüntüleme • 8 gün önce