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.

For years, I’ve been tuning parameters for robot designs and controllers on specific tasks. Now we can automate this on dataset-scale. Introducing Co-Design of Soft Gripper with Neural Physics - a soft gripper trained in simulation to deform while handling load.

yisha

1,215 subscribers

36,023 views • 1 year ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

10 Comments

yisha1 year ago

We derived a uniform pressure model for tendon-driven soft fingers. Instead of using only the fingertips, it utilizes the entire finger by deforming to any shape.

yisha1 year ago

webpage: paper:

yisha1 year ago

This is done with my amazing collaborators @baicrystal25, Adabhav Singh, @jianglong_ye , Mike Tolley, @xiaolonw

yisha1 year ago

✨We show that jointly optimizing on all data performs better than individually optimized grippers✨

Tony Tao1 year ago

Cool work Yi Sha!

yisha1 year ago

Thank you ☺️

Sarvesh Patil1 year ago

This is really cool Yisha!!! Did you find warp easy to use?

yisha1 year ago

It’s easier to debug because it’s open source. In general I feel like GPU+soft body simulation still has a long way to go 😂

Sarvesh Patil1 year ago

Haha makes sense! Cheers of the paper!

yisha1 year ago

We sample diverse design (stiffness) parameters and poses in simulation. We use this to train a neural network surrogate, and then optimize the best design and pose. And then transfer to real robots using structure stiffness in 3d printing.

Related Videos

Click-and-Drag Gripper Control is now LIVE We've been simplifying high-hardware-demand, complex simulation teleoperation through web-based keyboard-and-mouse control — and now we're taking it one step further. Direct click-and-drag gripper control is officially here. This eliminates a significant number of intermediate steps. Simply reorient your viewing plane by adjusting the camera perspective, and focus only on gripper-object contact. See it in action 👇

Click-and-Drag Gripper Control is now LIVE We've been simplifying high-hardware-demand, complex simulation teleoperation through web-based keyboard-and-mouse control — and now we're taking it one step further. Direct click-and-drag gripper control is officially here. This eliminates a significant number of intermediate steps. Simply reorient your viewing plane by adjusting the camera perspective, and focus only on gripper-object contact. See it in action 👇

Axis Robotics

30,887 views • 3 months ago

🔥 #ICRA2026 Best Paper Finalist The era of "robot VLA = single-arm gripper" is ending. Introducing Dexora — the first open-source Vision-Language-Action system for dual-arm, dual-hand, 36-DoF dexterous manipulation. 🦾 Dual Arms 🖐️ Dual Hands 🎯 36 DoF Control 🌍 Open Source Trained on: • 100K simulated trajectories • 10K real-world demonstrations Dexora achieves: ✓ 90%+ success on basic manipulation ✓ Strong dexterous manipulation performance ✓ Cross-embodiment generalization Our key hypothesis: Train on the hardest embodiment. Transfer to simpler robots later. Instead of scaling up gripper policies, we train directly in the most expressive action space and project downward to simpler embodiments. This may be a practical path toward universal robot controllers. 🎥 Demos: 📄 Paper:

🔥 #ICRA2026 Best Paper Finalist The era of "robot VLA = single-arm gripper" is ending. Introducing Dexora — the first open-source Vision-Language-Action system for dual-arm, dual-hand, 36-DoF dexterous manipulation. 🦾 Dual Arms 🖐️ Dual Hands 🎯 36 DoF Control 🌍 Open Source Trained on: • 100K simulated trajectories • 10K real-world demonstrations Dexora achieves: ✓ 90%+ success on basic manipulation ✓ Strong dexterous manipulation performance ✓ Cross-embodiment generalization Our key hypothesis: Train on the hardest embodiment. Transfer to simpler robots later. Instead of scaling up gripper policies, we train directly in the most expressive action space and project downward to simpler embodiments. This may be a practical path toward universal robot controllers. 🎥 Demos: 📄 Paper:

Hao Zhao

17,474 views • 3 months ago

ANIME PHYSICS IN BLENDER After years of R&D, we're finally releasing our custom-built physics solution for Blender! Introducing: Goo Physics Goo Physics brings bouncy, floppy, and gooey stylized bone simulation to your rigs with a single click of a button. Available now at ➡️

ANIME PHYSICS IN BLENDER After years of R&D, we're finally releasing our custom-built physics solution for Blender! Introducing: Goo Physics Goo Physics brings bouncy, floppy, and gooey stylized bone simulation to your rigs with a single click of a button. Available now at ➡️

DillonGoo Studios

4,014,805 views • 5 months ago

Introducing Axis Dataset V1 - the simulation dataset for scalable robot manipulation. Can noisy, crowdsourced simulation data support embodied pretraining? Yes—with enough coverage and diversity. Continual pretraining on AXIS dataset V1 lifts π0.5 from 83.9% to 88.8% on LIBERO-Plus, with performance improving consistently as the pretraining data scales from 25% to 100% of the full dataset, showing no clear saturation. Built with researchers from Georgia Tech UC Berkeley Texas A&M University Johns Hopkins University Penn University of Michigan NUS NTU Singapore. ➡️Related links: Paper: Dataset: Project Page: GitHub Codebase:

Introducing Axis Dataset V1 - the simulation dataset for scalable robot manipulation. Can noisy, crowdsourced simulation data support embodied pretraining? Yes—with enough coverage and diversity. Continual pretraining on AXIS dataset V1 lifts π0.5 from 83.9% to 88.8% on LIBERO-Plus, with performance improving consistently as the pretraining data scales from 25% to 100% of the full dataset, showing no clear saturation. Built with researchers from Georgia Tech UC Berkeley Texas A&M University Johns Hopkins University Penn University of Michigan NUS NTU Singapore. ➡️Related links: Paper: Dataset: Project Page: GitHub Codebase:

Axis Robotics

96,538 views • 1 month ago

For the past few months, we’ve been building something I’ve always wished existed: Introducing Wonder - the first AI-native design tool on an infinite canvas with taste and understanding of your designs.

For the past few months, we’ve been building something I’ve always wished existed: Introducing Wonder - the first AI-native design tool on an infinite canvas with taste and understanding of your designs.

Aibek Yegemberdin

902,054 views • 1 year ago

Introducing SimReadyGen. Agentic simulation generation for Physical AI. From a text prompt to physically accurate SimReady assets at scale. Robot training demands physically accurate environments at a scale traditional asset creation can't match. SimReadyGen closes that gap. Every generated asset carries measured physics, not estimates. Behind it is SimReady Foundry, our measured-physics pipeline: we measure how real objects behave — contact, friction, dynamics — and turn those measurements into ground-truth simulation parameters. Built on OpenUSD and integrated with NVIDIA Robotics Omniverse Libraries, SimReadyGen moves directly into Isaac Sim and Isaac Lab workflows. Try it today: Read the announcement: #PhysicalAI #Robotics #Simulation #OpenUSD

Introducing SimReadyGen. Agentic simulation generation for Physical AI. From a text prompt to physically accurate SimReady assets at scale. Robot training demands physically accurate environments at a scale traditional asset creation can't match. SimReadyGen closes that gap. Every generated asset carries measured physics, not estimates. Behind it is SimReady Foundry, our measured-physics pipeline: we measure how real objects behave — contact, friction, dynamics — and turn those measurements into ground-truth simulation parameters. Built on OpenUSD and integrated with NVIDIA Robotics Omniverse Libraries, SimReadyGen moves directly into Isaac Sim and Isaac Lab workflows. Try it today: Read the announcement: #PhysicalAI #Robotics #Simulation #OpenUSD

Lightwheel

12,221 views • 2 months ago

In my past research experience, finding or developing an appropriate simulation environment, dataset, and benchmark has always been a challenge. Missing features, limited support, or unexpected bugs often occupied my days and nights. Moreover, current simulation platforms are relatively fragmented—making it challenging to replicate the success of the RT-X dataset in unifying community efforts. Introducing RoboVerse, we provide a unified platform, dataset, and benchmark for scalable and generalizable robot learning. We hope to build a shared foundation to combine the community efforts. RoboVerse includes: MetaSim: We carefully designed a configuration system and a universal interface to align current robotic simulators. With MetaSim, you can use any simulator with the same code—bringing together the community’s diverse efforts under one framework! RoboVerse Dataset and Benchmark: We unify popular simulation environments and benchmarks into a single cohesive system and introduce the RoboVerse dataset—a large-scale, high-quality synthetic dataset. Additionally, we propose a standardized benchmark across both imitation learning and reinforcement learning. A cool feature enabled by our unified framework: Hybrid Simulation! You can now integrate physics engines and renderers from different simulators—e.g., using MuJoCo precise physics with Isaac photorealistic rendering. This not only elevates simulation fidelity but also significantly enhances real-world transfer performance across complex robotic applications. Hopefully, our team’s efforts could serve the robotic community to thrive vibrantly in the years to come. RoboVerse is open-sourced🥳!!! Project Page: Documentation: Github Repo: Paper:

In my past research experience, finding or developing an appropriate simulation environment, dataset, and benchmark has always been a challenge. Missing features, limited support, or unexpected bugs often occupied my days and nights. Moreover, current simulation platforms are relatively fragmented—making it challenging to replicate the success of the RT-X dataset in unifying community efforts. Introducing RoboVerse, we provide a unified platform, dataset, and benchmark for scalable and generalizable robot learning. We hope to build a shared foundation to combine the community efforts. RoboVerse includes: MetaSim: We carefully designed a configuration system and a universal interface to align current robotic simulators. With MetaSim, you can use any simulator with the same code—bringing together the community’s diverse efforts under one framework! RoboVerse Dataset and Benchmark: We unify popular simulation environments and benchmarks into a single cohesive system and introduce the RoboVerse dataset—a large-scale, high-quality synthetic dataset. Additionally, we propose a standardized benchmark across both imitation learning and reinforcement learning. A cool feature enabled by our unified framework: Hybrid Simulation! You can now integrate physics engines and renderers from different simulators—e.g., using MuJoCo precise physics with Isaac photorealistic rendering. This not only elevates simulation fidelity but also significantly enhances real-world transfer performance across complex robotic applications. Hopefully, our team’s efforts could serve the robotic community to thrive vibrantly in the years to come. RoboVerse is open-sourced🥳!!! Project Page: Documentation: Github Repo: Paper:

Haoran Geng

84,318 views • 1 year ago

Major milestone in Reinforcement Learning 🚀 In under just 1 hour of training, our AI Agent can now automate tasks requiring complex physics and object interactions. Done for the first time on 100$ robot arms. Watch as it recovers when the cube gets stuck 🤯

Major milestone in Reinforcement Learning 🚀 In under just 1 hour of training, our AI Agent can now automate tasks requiring complex physics and object interactions. Done for the first time on 100$ robot arms. Watch as it recovers when the cube gets stuck 🤯

Remi Cadene

62,536 views • 1 year ago

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Jim Fan

467,237 views • 1 year ago

Let's reverse engineer this demo. You need 3 things: (1) robust hardware and motor designs that treat simulation as first-class citizen; (2) a human motion capture ("mocap") dataset, such as those for film and gaming characters; (3) massively parallel RL training in GPU-accelerated simulation. Last October, our team trained a 1.5M parameter foundation model called HOVER for such agile motor control. It follows this recipe, roughly speaking (details in thread): (1) Simulation used to be an after-thought. Now, it has to be part of the hardware design process. If your robot doesn't simulate well, you can kiss RL goodbye. Hardware-simulation co-design is a very interesting emergent topic that only becomes meaningful with today's compute capability. (2) Human mocap dataset to produce natural-looking walking and running gaits. That's one huge advantage of using humanoid robot - you get to imitate from tons of human motions that were originally captured for movies or AAA games. At least 3 ways to use the data: - For initialization: pre-train the neural net to imitate human, and then finetune it into the robot form factor with physics turned on; - For reward function: penalize any deviations from the target pose; - For representation learning: treat the human poses as a "motion prior" to constrain the space of robot behaviors. (3) Shove the above into Isaac sim, add a lot of randomization, pump it through PPO, throw in a bunch of GPUs, and then watch Netflix till loss converges. If you have an urge to comment this is CGI, let me save you a few keystrokes — many academic labs now own the G1 robot in the flesh. See our team's HOVER work in the thread: 🧵

Let's reverse engineer this demo. You need 3 things: (1) robust hardware and motor designs that treat simulation as first-class citizen; (2) a human motion capture ("mocap") dataset, such as those for film and gaming characters; (3) massively parallel RL training in GPU-accelerated simulation. Last October, our team trained a 1.5M parameter foundation model called HOVER for such agile motor control. It follows this recipe, roughly speaking (details in thread): (1) Simulation used to be an after-thought. Now, it has to be part of the hardware design process. If your robot doesn't simulate well, you can kiss RL goodbye. Hardware-simulation co-design is a very interesting emergent topic that only becomes meaningful with today's compute capability. (2) Human mocap dataset to produce natural-looking walking and running gaits. That's one huge advantage of using humanoid robot - you get to imitate from tons of human motions that were originally captured for movies or AAA games. At least 3 ways to use the data: - For initialization: pre-train the neural net to imitate human, and then finetune it into the robot form factor with physics turned on; - For reward function: penalize any deviations from the target pose; - For representation learning: treat the human poses as a "motion prior" to constrain the space of robot behaviors. (3) Shove the above into Isaac sim, add a lot of randomization, pump it through PPO, throw in a bunch of GPUs, and then watch Netflix till loss converges. If you have an urge to comment this is CGI, let me save you a few keystrokes — many academic labs now own the G1 robot in the flesh. See our team's HOVER work in the thread: 🧵

Jim Fan

216,139 views • 1 year ago

Most motion papers tailor one controller to one specific task. This year at SIGGRAPH, our research team asks: can motor control itself be pretrained and reused? Generative Pretrained Controllers, or GPC, turn motor skills into a vocabulary of discrete tokens and train a transformer-based generative controller through next-token prediction. Just like GPT, the same pretrained controller can then be fine-tuned to solve new tasks. Trained on 600+ hours of motion, GPC runs in real-time inside a physics simulation, producing natural and physically grounded behaviors for interactive control.

Most motion papers tailor one controller to one specific task. This year at SIGGRAPH, our research team asks: can motor control itself be pretrained and reused? Generative Pretrained Controllers, or GPC, turn motor skills into a vocabulary of discrete tokens and train a transformer-based generative controller through next-token prediction. Just like GPT, the same pretrained controller can then be fine-tuned to solve new tasks. Trained on 600+ hours of motion, GPC runs in real-time inside a physics simulation, producing natural and physically grounded behaviors for interactive control.

NVIDIA AI

176,214 views • 2 months ago

New project in collaboration with Adrian Wilkins ! 🤖 It’s been a while since I’ve worked on a sci-fi project, so I’m excited to dive back into it. I’m handling the animation & rigging for this robot. More to come! #GameDev #GameAnimation #SciFi #robot #animation

New project in collaboration with Adrian Wilkins ! 🤖 It’s been a while since I’ve worked on a sci-fi project, so I’m excited to dive back into it. I’m handling the animation & rigging for this robot. More to come! #GameDev #GameAnimation #SciFi #robot #animation

Arthur Munoz

14,104 views • 27 days ago

Can robot foundation models scale? Xiaomi-Robotics-1 explores this question with over 100,000 hours of real-world manipulation data. Pre-trained on large-scale real-world trajectories and post-trained with cross-embodiment robot data, Xiaomi-Robotics-1 demonstrates consistent scaling across data and model size, strong generalization in unseen environments, and efficient adaptation to new tasks. 🔗 #Robotics #EmbodiedAI #FoundationModels #RobotLearning #XiaomiAI #XiaomiRobotics

Can robot foundation models scale? Xiaomi-Robotics-1 explores this question with over 100,000 hours of real-world manipulation data. Pre-trained on large-scale real-world trajectories and post-trained with cross-embodiment robot data, Xiaomi-Robotics-1 demonstrates consistent scaling across data and model size, strong generalization in unseen environments, and efficient adaptation to new tasks. 🔗 #Robotics #EmbodiedAI #FoundationModels #RobotLearning #XiaomiAI #XiaomiRobotics

Xiaomi Tech

23,600 views • 2 months ago

The World Model as NEO's Cognitive Core 1X has revealed a major AI development where the NEO humanoid can translate any natural language prompt into robotic action. It demonstrates this capability even for novel tasks, objects, and environments not found in its robot dataset. - the 1X World Model is trained on internet-scale human interaction videos and fine-tuned with robot data to ground its understanding in physics and in NEO's embodiment - from a simple voice or text prompt, the world model generates a visualization of future actions - a built-in inverse dynamics model then translates these into precise motor movements for NEO

The World Model as NEO's Cognitive Core 1X has revealed a major AI development where the NEO humanoid can translate any natural language prompt into robotic action. It demonstrates this capability even for novel tasks, objects, and environments not found in its robot dataset. - the 1X World Model is trained on internet-scale human interaction videos and fine-tuned with robot data to ground its understanding in physics and in NEO's embodiment - from a simple voice or text prompt, the world model generates a visualization of future actions - a built-in inverse dynamics model then translates these into precise motor movements for NEO

The Humanoid Hub

68,453 views • 8 months ago

I was really impressed by the UMI gripper (Cheng Chi et al.), but a key limitation is that **force-related data wasn’t captured**: humans feel haptic feedback through the mechanical springs, but the robot couldn’t leverage that info, limiting the data’s value for fine-grained manipulation tasks. Led by my amazing students Yolanda Zhu and Binghao Huang, we designed a **portable visuo-tactile gripper** by integrating our dense, flexible tactile arrays with the UMI gripper to enable large-scale in-the-wild data collection. 🔗 We demonstrate **cross-modal representation learning** and **downstream policy learning** on tasks requiring in-hand state estimation (e.g., test tube reorientation) and fine-grained force sensing (e.g., pipette fluid transfer). Key takeaways: - Our flexible tactile arrays store the rich haptic information humans perceive as dense tactile signals. - Portability and robustness are key for in-the-wild data collection; our portable gripper is compact, lightweight, and durable. - Touch provides precise, robust measurements of in-hand object pose, invariant to lighting and viewpoint. - Cross-modal pretraining on large-scale in-the-wild data significantly improves policy robustness and sample efficiency (as shown many times before — and verified again here!). Also check out our previous investigations of dense, flexible tactile grids for understanding human-robot-environment interactions: - Dense tactile glove (Nature ’19): - 3D-ViTac (CoRL ’24):

I was really impressed by the UMI gripper (Cheng Chi et al.), but a key limitation is that force-related data wasn’t captured: humans feel haptic feedback through the mechanical springs, but the robot couldn’t leverage that info, limiting the data’s value for fine-grained manipulation tasks. Led by my amazing students Yolanda Zhu and Binghao Huang, we designed a portable visuo-tactile gripper by integrating our dense, flexible tactile arrays with the UMI gripper to enable large-scale in-the-wild data collection. 🔗 We demonstrate cross-modal representation learning and downstream policy learning on tasks requiring in-hand state estimation (e.g., test tube reorientation) and fine-grained force sensing (e.g., pipette fluid transfer). Key takeaways: - Our flexible tactile arrays store the rich haptic information humans perceive as dense tactile signals. - Portability and robustness are key for in-the-wild data collection; our portable gripper is compact, lightweight, and durable. - Touch provides precise, robust measurements of in-hand object pose, invariant to lighting and viewpoint. - Cross-modal pretraining on large-scale in-the-wild data significantly improves policy robustness and sample efficiency (as shown many times before — and verified again here!). Also check out our previous investigations of dense, flexible tactile grids for understanding human-robot-environment interactions: - Dense tactile glove (Nature ’19): - 3D-ViTac (CoRL ’24):

Yunzhu Li

13,188 views • 1 year ago

DOBOT robotics' humanoid robot is making a big leap in industrial applications, achieving cross-scenario, multi-task generalization. This is thanks to two major breakthroughs: efficient human-to-robot motion mapping and knowledge-driven generative VLA tech. ► The robot handles complex tasks like precision assembly and can work reliably in temperatures over 50°C (122°F). ► It excels at multi-robot collaborative tasks and can adaptively grasp soft or irregularly shaped objects. With a repeatability of ±0.05mm, this model demonstrates strong adaptability and reliability in dynamic industrial settings. It's now being applied in warehouse anomaly handling and quality inspection.

DOBOT robotics' humanoid robot is making a big leap in industrial applications, achieving cross-scenario, multi-task generalization. This is thanks to two major breakthroughs: efficient human-to-robot motion mapping and knowledge-driven generative VLA tech. ► The robot handles complex tasks like precision assembly and can work reliably in temperatures over 50°C (122°F). ► It excels at multi-robot collaborative tasks and can adaptively grasp soft or irregularly shaped objects. With a repeatability of ±0.05mm, this model demonstrates strong adaptability and reliability in dynamic industrial settings. It's now being applied in warehouse anomaly handling and quality inspection.

RoboHub🤖

93,862 views • 1 year ago

I trained a grasping policy on objects of various shapes and sizes using the new mjlab feature and this really cool pivot grasp strategy emerges for large flat objects that are larger than the gripper aperture. So cool and beautiful to see!

I trained a grasping policy on objects of various shapes and sizes using the new mjlab feature and this really cool pivot grasp strategy emerges for large flat objects that are larger than the gripper aperture. So cool and beautiful to see!

Kevin Zakka

20,458 views • 4 months ago