Загрузка видео...

Не удалось загрузить видео

На главную

Introduce HumanPlus - Autonomous Skills part Humanoids are born for using human data. Imitating humans, our humanoid learns: - fold sweatshirts - unload objects from warehouse racks - diverse locomotion skills (squatting, jumping, standing) - greet another robot Open-sourced!

zipengfu's profile picture

Zipeng Fu

13,211 subscribers

158,373 просмотров • 2 лет назад •via X (Twitter)

ОбразованиеНаука и технологииИгры
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 10

Фото профиля Zipeng Fu
Zipeng Fu2 лет назад

We build our customized 33-DoF humanoid, and a data collection pipeline through real-time shadowing in the real world.

Фото профиля Zipeng Fu
Zipeng Fu2 лет назад

Using the data collected through shadowing, we then perform supervised behavior cloning to train skill policies using egocentric vision. We introduce Humanoid Imitation Transformer. Based on ACT, HIT adds forward dynamics prediction on image feature space as a regularization.

Фото профиля Zipeng Fu
Zipeng Fu2 лет назад

Compared to baselines, HIT uses - binocular vision, thus having implicit stereos for depth information - visual feedback better, avoiding overfitting to proprioception given small-sized demos

Фото профиля Zipeng Fu
Zipeng Fu2 лет назад

Besides vision-based whole-body manipulation skills, our humanoid has strong locomotion skills: - outperforming H1 default standing controller under strong perturbation forces - enabling more whole-body skills like squatting and jumping

Фото профиля Zipeng Fu
Zipeng Fu2 лет назад

This project is not possible without our team of experts, covering from computer graphics to robot learning to robot hardware: - co-leads: @qingqing_zhao_ @Qi_Wu577 - advisors: @chelseabfinn @GordonWetzstein project website: hardware: code:

Фото профиля Karol Hausman
Karol Hausman2 лет назад

Congrats @zipengfu !! 🎉

Фото профиля Yu Yang
Yu Yang2 лет назад

Pretty cool!

Фото профиля Joe Valentine
Joe Valentine2 лет назад

@DrDisrespect robodoc?

Фото профиля BotBot
BotBot2 лет назад

Wow! Incredible, is the demo unit the Unitree H1?

Фото профиля Ak
Ak2 лет назад

@heyjchu

Похожие видео

Introduce HumanPlus - Shadowing part Humanoids are born for using human data. We build a real-time shadowing system using a single RGB camera and a whole-body policy for cloning human motion. Examples: - boxing🥊 - playing the piano🎹/ping pong - tossing - typing Open-sourced!
0:53

Introduce HumanPlus - Shadowing part Humanoids are born for using human data. We build a real-time shadowing system using a single RGB camera and a whole-body policy for cloning human motion. Examples: - boxing🥊 - playing the piano🎹/ping pong - tossing - typing Open-sourced!

Zipeng Fu

230,388 просмотров • 2 лет назад

Excited to introduce COLA: Learning Human–Humanoid Coordination for Collaborative Object Carrying 🤝🤖 COLA makes humanoids truly helpful in human collaboration — capable of carrying objects, pushing carts, or responding to human push commands. It provides a proprioception-only policy for compliant human–humanoid coordination across diverse movement patterns. The core idea is simple yet effective: 👉 Fine-tune a collaborative policy from a locomotion policy using a residual teacher. 👉 Train in simulation and distill to a real-world student policy for deployment. Paper: Project:
3:13

Excited to introduce COLA: Learning Human–Humanoid Coordination for Collaborative Object Carrying 🤝🤖 COLA makes humanoids truly helpful in human collaboration — capable of carrying objects, pushing carts, or responding to human push commands. It provides a proprioception-only policy for compliant human–humanoid coordination across diverse movement patterns. The core idea is simple yet effective: 👉 Fine-tune a collaborative policy from a locomotion policy using a residual teacher. 👉 Train in simulation and distill to a real-world student policy for deployment. Paper: Project:

Siyuan Huang

22,030 просмотров • 11 месяцев назад

Another American industrial humanoid robot🤖🏭 Noble Machines (formerly UCR) has just emerged from stealth mode. Launching the first humanoid robot. Founded only 18 months ago, it has already been deployed in a Fortune Global 500 factory. Teams from SpaceX, NASA, and Apple specialize in the toughest, most tiring, and most dangerous industrial tasks: >27kg heavy load >5-hour battery life >Walking speed 0.8m/s >Climbing stairs, traversing scaffolding, and navigating chaotic construction sites > Modular end effector, allowing for quick tool change. > AI-controlled operation with end-to-end autonomy; learns new skills in hours >Autonomous operation + Telep-op mode >Rapid integration with existing enterprise workflows > Human-robot collaboration
1:12

Another American industrial humanoid robot🤖🏭 Noble Machines (formerly UCR) has just emerged from stealth mode. Launching the first humanoid robot. Founded only 18 months ago, it has already been deployed in a Fortune Global 500 factory. Teams from SpaceX, NASA, and Apple specialize in the toughest, most tiring, and most dangerous industrial tasks: >27kg heavy load >5-hour battery life >Walking speed 0.8m/s >Climbing stairs, traversing scaffolding, and navigating chaotic construction sites > Modular end effector, allowing for quick tool change. > AI-controlled operation with end-to-end autonomy; learns new skills in hours >Autonomous operation + Telep-op mode >Rapid integration with existing enterprise workflows > Human-robot collaboration

CyberRobo

44,656 просмотров • 7 месяцев назад

Humanoids finally move like humans… and can do more than copy. [Details + demos in thread 👇] A new framework, BeyondMimic, shows how to learn naturalistic whole-body control from human motion. But then goes further by composing those skills into versatile, zero-shot policies. Two key pieces make it work: •Motion tracking pipeline: robustly turns human mocap into dynamic robot motions (jump spins, sprinting, cartwheels) on real hardware •Guided diffusion policy: distills motion primitives and enables zero-shot downstream tasks with simple cost functions The result: ✅ State-of-the-art motion quality from mocap tracking ✅ Versatile task execution; waypoint navigation, joystick teleop, obstacle avoidance ✅ Stable sim-to-real transfer for diverse humanoid skills Thank you for sharing, Qiayuan Liao! 📍 Paper: Video:
0:42

Humanoids finally move like humans… and can do more than copy. [Details + demos in thread 👇] A new framework, BeyondMimic, shows how to learn naturalistic whole-body control from human motion. But then goes further by composing those skills into versatile, zero-shot policies. Two key pieces make it work: •Motion tracking pipeline: robustly turns human mocap into dynamic robot motions (jump spins, sprinting, cartwheels) on real hardware •Guided diffusion policy: distills motion primitives and enables zero-shot downstream tasks with simple cost functions The result: ✅ State-of-the-art motion quality from mocap tracking ✅ Versatile task execution; waypoint navigation, joystick teleop, obstacle avoidance ✅ Stable sim-to-real transfer for diverse humanoid skills Thank you for sharing, Qiayuan Liao! 📍 Paper: Video:

Ilir Aliu - eu/acc

72,522 просмотров • 1 год назад

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
1:54

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

118,521 просмотров • 1 год назад

Humanoid robots are starting to build humanoid robots. UBTECH just put a 14,000㎡ humanoid robot factory into operation in Liuzhou, China, with a designed takt time of one humanoid every 10 minutes and planned annual capacity in the tens of thousands. What’s interesting is that humanoid robots are already part of the production process. Cruzr Y1 and Y2 handle depalletizing, palletizing, material feeding and transport, using 3D vision to adapt to shifted boxes and changing pallet patterns. On the assembly side, Walker S2 and Cruzr are produced on the same line alongside cobots, autonomous logistics vehicles and assistive manipulators. The factory also has a 65㎡ automated warehouse capable of storing 112 humanoids, with AGVs, industrial robots and stacker cranes coordinated as one system. Before production, the factory was modeled 1:1 in Siemens Plant Simulation, covering workstations, racks and AGVs. Each robot gets its own SN, with components, batches, assembly parameters and inspection data tracked all the way to delivery. Humanoid robots are now entering automated manufacturing at the 10,000-unit scale. The next milestone may be when they can take part in enough of the assembly, testing and production process to truly build robots themselves.
1:02

Humanoid robots are starting to build humanoid robots. UBTECH just put a 14,000㎡ humanoid robot factory into operation in Liuzhou, China, with a designed takt time of one humanoid every 10 minutes and planned annual capacity in the tens of thousands. What’s interesting is that humanoid robots are already part of the production process. Cruzr Y1 and Y2 handle depalletizing, palletizing, material feeding and transport, using 3D vision to adapt to shifted boxes and changing pallet patterns. On the assembly side, Walker S2 and Cruzr are produced on the same line alongside cobots, autonomous logistics vehicles and assistive manipulators. The factory also has a 65㎡ automated warehouse capable of storing 112 humanoids, with AGVs, industrial robots and stacker cranes coordinated as one system. Before production, the factory was modeled 1:1 in Siemens Plant Simulation, covering workstations, racks and AGVs. Each robot gets its own SN, with components, batches, assembly parameters and inspection data tracked all the way to delivery. Humanoid robots are now entering automated manufacturing at the 10,000-unit scale. The next milestone may be when they can take part in enough of the assembly, testing and production process to truly build robots themselves.

CyberRobo

17,883 просмотров • 21 дней назад

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: 🧵
2:21

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

467,416 просмотров • 1 год назад

Sports like tennis are great examples of the sort of dynamic whole-body interaction that’s possible with humanoid robots. But capturing examples of fast, dynamic interactions from humans is really difficult. Enter LATENT, which uses lower-quality human data plus reinforcement learning to teach a robot to play tennis, able to complete back-and-forth volleys at a human level. LATENT has three steps: (1) collecting imperfect human data like a backswing, (2) using these to learn a latent action space, and (3) they train a high-level policy in simulation which can compose these actions and execute tennis skills on a robot. Haofei Lu and Yunrui Lian ✈️ IROS join us to tell us about their method. Watch Episode #80 of RoboPapers, with Chris Paxton and Jiafei Duan, now to learn more!
55:30

Sports like tennis are great examples of the sort of dynamic whole-body interaction that’s possible with humanoid robots. But capturing examples of fast, dynamic interactions from humans is really difficult. Enter LATENT, which uses lower-quality human data plus reinforcement learning to teach a robot to play tennis, able to complete back-and-forth volleys at a human level. LATENT has three steps: (1) collecting imperfect human data like a backswing, (2) using these to learn a latent action space, and (3) they train a high-level policy in simulation which can compose these actions and execute tennis skills on a robot. Haofei Lu and Yunrui Lian ✈️ IROS join us to tell us about their method. Watch Episode #80 of RoboPapers, with Chris Paxton and Jiafei Duan, now to learn more!

RoboPapers

30,095 просмотров • 4 месяцев назад

I don’t know if we live in a Matrix, but I know for sure that robots will spend most of their lives in simulation. Let machines train machines. I’m excited to introduce DexMimicGen, a massive-scale synthetic data generator that enables a humanoid robot to learn complex skills from only a handful of human demonstrations. Yes, as few as 5! DexMimicGen addresses the biggest pain point in robotics: where do we get data? Unlike with LLMs, where vast amounts of texts are readily available, you cannot simply download motor control signals from the internet. So researchers teleoperate the robots to collect motion data via XR headsets. They have to repeat the same skill over and over and over again, because neural nets are data hungry. This is a very slow and uncomfortable process. At NVIDIA, we believe the majority of high-quality tokens for robot foundation models will come from simulation. What DexMimicGen does is to trade GPU compute time for human time. It takes one motion trajectory from human, and multiplies into 1000s of new trajectories. A robot brain trained on this augmented dataset will generalize far better in the real world. Think of DexMimicGen as a learning signal amplifier. It maps a small dataset to a large (de facto infinite) dataset, using physics simulation in the loop. In this way, we free humans from babysitting the bots all day. The future of robot data is generative. The future of the entire robot learning pipeline will also be generative. 🧵
0:58

I don’t know if we live in a Matrix, but I know for sure that robots will spend most of their lives in simulation. Let machines train machines. I’m excited to introduce DexMimicGen, a massive-scale synthetic data generator that enables a humanoid robot to learn complex skills from only a handful of human demonstrations. Yes, as few as 5! DexMimicGen addresses the biggest pain point in robotics: where do we get data? Unlike with LLMs, where vast amounts of texts are readily available, you cannot simply download motor control signals from the internet. So researchers teleoperate the robots to collect motion data via XR headsets. They have to repeat the same skill over and over and over again, because neural nets are data hungry. This is a very slow and uncomfortable process. At NVIDIA, we believe the majority of high-quality tokens for robot foundation models will come from simulation. What DexMimicGen does is to trade GPU compute time for human time. It takes one motion trajectory from human, and multiplies into 1000s of new trajectories. A robot brain trained on this augmented dataset will generalize far better in the real world. Think of DexMimicGen as a learning signal amplifier. It maps a small dataset to a large (de facto infinite) dataset, using physics simulation in the loop. In this way, we free humans from babysitting the bots all day. The future of robot data is generative. The future of the entire robot learning pipeline will also be generative. 🧵

Jim Fan

165,246 просмотров • 1 год назад

Another headless wheeled humanoid Shenzhen VLAI Robotics just released a pretty pragmatic humanoid robot called L1. I checked the price, starting at RMB 28,800($4,188). wheeled base + height-adjustable dual arms.Anyone can be built into a household assistant:fetching water, tidying up the kitchen and living room, doing laundry, and more. >Dual arms with 16 DOF, bionic-style structure, capable of relatively fine tasks like grasping and plug-in operations >Adjustable arm height (~70–160 cm), covering both tabletop and standing work >Differential-drive wheeled base, 100 kg payload, simple but well-suited for indoor use, supports working while moving >Open interfaces + teleoperation/data collection. L1 feels more like a development platform than a finished product Beyond household humanoids, developers can build specialized humanoid robots tailored to a wide range of scenarios,from scientific research and manufacturing to warehousing, retail, and caregiving... We’re already seeing quite a few robots in this form factor. While some are pushing toward full humanoids, this “less-human” approach might actually reach real-world deployment sooner. --- (Shenzhen truly is a home for humanoid robots,it seems as though a new humanoid robot company could spring into existence at any moment.）
2:02

Another headless wheeled humanoid Shenzhen VLAI Robotics just released a pretty pragmatic humanoid robot called L1. I checked the price, starting at RMB 28,800($4,188). wheeled base + height-adjustable dual arms.Anyone can be built into a household assistant:fetching water, tidying up the kitchen and living room, doing laundry, and more. >Dual arms with 16 DOF, bionic-style structure, capable of relatively fine tasks like grasping and plug-in operations >Adjustable arm height (~70–160 cm), covering both tabletop and standing work >Differential-drive wheeled base, 100 kg payload, simple but well-suited for indoor use, supports working while moving >Open interfaces + teleoperation/data collection. L1 feels more like a development platform than a finished product Beyond household humanoids, developers can build specialized humanoid robots tailored to a wide range of scenarios,from scientific research and manufacturing to warehousing, retail, and caregiving... We’re already seeing quite a few robots in this form factor. While some are pushing toward full humanoids, this “less-human” approach might actually reach real-world deployment sooner. --- (Shenzhen truly is a home for humanoid robots,it seems as though a new humanoid robot company could spring into existence at any moment.）

CyberRobo

38,700 просмотров • 6 месяцев назад

Excited to share our latest work on 🎧spatial audio-driven human motion generation. We aim to tackle a largely underexplored yet important problem of enabling virtual humans to move naturally in response to spatial audio—capturing not just what is heard, but also where the sound is coming from. To this end, we introduce the Spatial Audio-Driven Human Motion (SAM) dataset—the first comprehensive dataset featuring paired high-quality human motion and spatial audio recordings. For benchmarking, we develop a generative framework for human MOtion generation driven by SPAtial audio, termed MOSPA, which learns to synthesize realistic and diverse human motions conditioned on spatial audio input. We hope this research could provide a foundation for future research in spatial perception, virtual characters, and embodied AI. The dataset and model will be open-sourced soon. A big thank you to our intern, Shuyang Xu, for the wonderful collaboration! Congratulations, Shuyang! Project page: Paper: Video: #Animation #CG #CV #AIGC #DL #Deeplearning #Motion #Graphics #AI #GenerativeAI
3:58

Excited to share our latest work on 🎧spatial audio-driven human motion generation. We aim to tackle a largely underexplored yet important problem of enabling virtual humans to move naturally in response to spatial audio—capturing not just what is heard, but also where the sound is coming from. To this end, we introduce the Spatial Audio-Driven Human Motion (SAM) dataset—the first comprehensive dataset featuring paired high-quality human motion and spatial audio recordings. For benchmarking, we develop a generative framework for human MOtion generation driven by SPAtial audio, termed MOSPA, which learns to synthesize realistic and diverse human motions conditioned on spatial audio input. We hope this research could provide a foundation for future research in spatial perception, virtual characters, and embodied AI. The dataset and model will be open-sourced soon. A big thank you to our intern, Shuyang Xu, for the wonderful collaboration! Congratulations, Shuyang! Project page: Paper: Video: #Animation #CG #CV #AIGC #DL #Deeplearning #Motion #Graphics #AI #GenerativeAI

Zhiyang (Frank) Dou

14,704 просмотров • 1 год назад

Robotics has a data scarcity problem - you simply can't scrape robot control data from webpages. Introducing GR00T-Mimic and GR00T-Gen: using both Graphics 1.0 & Graphics 2.0 to multiply your robot datasets by 1,000,000x. We trade compute for synthetic data, so we are not capped by the fundamental physical limit of 24 hrs/robot/day. Robotics is right in the thick of Moravec's paradox: things that are easy for humans turn out to be incredibly hard for machines. We are crushing the Moravec's paradox, one token at a time. > Graphics 1.0: Isaac simulators with manually written, GPU-accelerated physics and rendering equations. > Graphics 2.0: big neural nets (Cosmos) that repaint the pixels from sim textures to real, given an open-ended prompt. Robot data multiplier workflow: 1. GR00T-Teleop: use XR device like Apple Vision Pro to map human finger poses to humanoid hands. 2. GR00T-Mimic: given a human-collected task demonstration, we augment the actions in Isaac and filter out ones that fail the task. 3. GR00T-Gen: apply Graphics 1.0 and then Graphics 2.0 to produce tons of visual variations. Join us to build Physical AI together! Job application links in thread:
2:33

Robotics has a data scarcity problem - you simply can't scrape robot control data from webpages. Introducing GR00T-Mimic and GR00T-Gen: using both Graphics 1.0 & Graphics 2.0 to multiply your robot datasets by 1,000,000x. We trade compute for synthetic data, so we are not capped by the fundamental physical limit of 24 hrs/robot/day. Robotics is right in the thick of Moravec's paradox: things that are easy for humans turn out to be incredibly hard for machines. We are crushing the Moravec's paradox, one token at a time. > Graphics 1.0: Isaac simulators with manually written, GPU-accelerated physics and rendering equations. > Graphics 2.0: big neural nets (Cosmos) that repaint the pixels from sim textures to real, given an open-ended prompt. Robot data multiplier workflow: 1. GR00T-Teleop: use XR device like Apple Vision Pro to map human finger poses to humanoid hands. 2. GR00T-Mimic: given a human-collected task demonstration, we augment the actions in Isaac and filter out ones that fail the task. 3. GR00T-Gen: apply Graphics 1.0 and then Graphics 2.0 to produce tons of visual variations. Join us to build Physical AI together! Job application links in thread:

Jim Fan

163,670 просмотров • 1 год назад

Today, we're joined by Nikita Rudin, co-founder and CEO of Flexion to discuss the gap between current robotic capabilities and what’s required to deploy fully autonomous robots in the real world. Nikita explains how reinforcement learning and simulation have driven rapid progress in robot locomotion—and why locomotion is still far from “solved.” We dig into the sim2real gap, and how adding visual inputs introduces noise and significantly complicates sim-to-real transfer. We also explore the debate between end-to-end models and modular approaches, and why separating locomotion, planning, and semantics remains a pragmatic approach today. Nikita also introduces the concept of "real-to-sim", which uses real-world data to refine simulation parameters for higher fidelity training, discusses how reinforcement learning, imitation learning, and teleoperation data are combined to train robust policies for both quadruped and humanoid robots, and introduces Flexion's hierarchical approach that utilizes pre-trained Vision-Language Models (VLMs) for high-level task orchestration with Vision-Language-Action (VLA) models and low-level whole-body trackers. Finally, Nikita shares the behind-the-scenes in humanoid robot demos, his take on reinforcement learning in simulation versus the real world, the nuances of reward tuning, and offers practical advice for researchers and practitioners looking to get started in robotics today. 🗒️ For the full list of resources for this episode, visit the show notes page: 📖 CHAPTERS =============================== 00:00 - Introduction 04:07 - Is robot locomotion solved? 06:04 - Sim-to-real gap 08:58 - Adding semantics to policies 09:42 - Modular vs end-to-end architectures 10:29 - Planner model 12:21 - Adapting RL techniques from quadrupeds to humanoids 15:39 - Behind robot demos 18:09 - Humanoid robots in home environments 22:03 - Training approach 23:56 - VLA models 27:59 - Closing the sim-to-real gap 32:55 - Task orchestration using VLMs 36:38 - Tool use 38:10 - Model hierarchy 43:37 - Simulator versus simulation environment 44:57 - Combining imitation learning and reinforcement learning 46:42 - RL in real world versus RL in simulation 52:58 - Reward tuning and value functions in robotics 56:38 - Predictions 1:00:10 - Humanoids, quadropeds, and wheeled platforms 1:02:45 - Advice, recommended robot kits, and community pla
1:06:07

Today, we're joined by Nikita Rudin, co-founder and CEO of Flexion to discuss the gap between current robotic capabilities and what’s required to deploy fully autonomous robots in the real world. Nikita explains how reinforcement learning and simulation have driven rapid progress in robot locomotion—and why locomotion is still far from “solved.” We dig into the sim2real gap, and how adding visual inputs introduces noise and significantly complicates sim-to-real transfer. We also explore the debate between end-to-end models and modular approaches, and why separating locomotion, planning, and semantics remains a pragmatic approach today. Nikita also introduces the concept of "real-to-sim", which uses real-world data to refine simulation parameters for higher fidelity training, discusses how reinforcement learning, imitation learning, and teleoperation data are combined to train robust policies for both quadruped and humanoid robots, and introduces Flexion's hierarchical approach that utilizes pre-trained Vision-Language Models (VLMs) for high-level task orchestration with Vision-Language-Action (VLA) models and low-level whole-body trackers. Finally, Nikita shares the behind-the-scenes in humanoid robot demos, his take on reinforcement learning in simulation versus the real world, the nuances of reward tuning, and offers practical advice for researchers and practitioners looking to get started in robotics today. 🗒️ For the full list of resources for this episode, visit the show notes page: 📖 CHAPTERS =============================== 00:00 - Introduction 04:07 - Is robot locomotion solved? 06:04 - Sim-to-real gap 08:58 - Adding semantics to policies 09:42 - Modular vs end-to-end architectures 10:29 - Planner model 12:21 - Adapting RL techniques from quadrupeds to humanoids 15:39 - Behind robot demos 18:09 - Humanoid robots in home environments 22:03 - Training approach 23:56 - VLA models 27:59 - Closing the sim-to-real gap 32:55 - Task orchestration using VLMs 36:38 - Tool use 38:10 - Model hierarchy 43:37 - Simulator versus simulation environment 44:57 - Combining imitation learning and reinforcement learning 46:42 - RL in real world versus RL in simulation 52:58 - Reward tuning and value functions in robotics 56:38 - Predictions 1:00:10 - Humanoids, quadropeds, and wheeled platforms 1:02:45 - Advice, recommended robot kits, and community pla

The TWIML AI Podcast

22,592 просмотров • 8 месяцев назад

New framework: Kick down your robot, it will get back up every time 🥋 Chinese startup RoboParty is a Beijing startup founded April 2025 by Huang Yi, originally shipping ROBOTO ORIGIN, the world's first full-stack open-source bipedal humanoid. They released UFO: Unsupervised Reinforcement Learning Framework for Humanoid Control. DEFINITIONS -> what differs is where the learning signal comes from: - SUPERVISED: humans supply the right answers (labels), the model imitates them. - UNSUPERVISED: no answer key, the model finds structure in raw data on its own. - REINFORCEMENT LEARNING: no answer key either, the model tries things and a reward scores each attempt. → UNSUPERVISED RL: trial and error where the agent invents its own rewards, instead of engineers hand-writing one per task. REPRESENTATION LEARNING: compress raw states into a useful internal map. TEMPORAL DISTANCE: distance on that map is "how many steps from A to B." CONTRASTIVE: trained by pulling together what's close in time, pushing apart what isn't. -> CONTRASTIVE TEMPORAL-DISTANCE REPRESENTATION LEARNING: the model builds an internal map of body states where distance means how many steps it takes to get from one to another. It is trained by contrast: states that occur close together in a movement get pulled together in the map, randomly paired states get pushed apart. UFO is an open-source training framework that teaches humanoid robots skills, like getting up, walking, goal-reaching, teleoperation, without reference motions -> no motion-capture or human-video demonstrations to imitate. Its core is TeCH, a contrastive temporal-distance representation-learning algorithm: the robot explores, builds pseudo-goals by temporal rolling, and learns goal-conditioned policies from a single unified progress reward. One framework trains five different robots (Unitree G1/H1, RoboParty RP0/RP1, AgiBot X2) with automatic config conversion in ~2–3 hours per robot! The real novelty here "no demonstrations at all". No data-collection arms race,the dominant humanoid-locomotion recipe is tracking: imitate mocap/retargeted-human reference trajectories. The robot self-generates goals from its own exploration and learns from a progress reward, needing zero reference motion data. Everybody else is fighting over data acquisition, while this team just teleports out of the race entirely (inb4 "competition is for losers 💀 ). This strategy reminds me of the DeepSeek playbook applied to robots: open-source the whole stack to become the global default and commoditize everyone else. RoboParty is giving away hardware and now control software (UFO) to be the Android of humanoids. Yet another reason for the US to ban Chinese open models perhaps 🥶 ? What I also really like about this approach is the cross-embodiment infrastructure, one framework trains Unitree G1/H1, RoboParty RP0/RP1, and AgiBot X2 with automatic configuration conversion. Just like Physical Intelligence, RoboParty seems to place itself as a neutral hardware agnostic middle man. Also woth mentioning: their ability ot perform stable skill injection, e.g. adding a cartwheel without forgetting how to walk. A common failure of RL humanoid policies is that teaching a new agile skill destabilizes the existing ones (catastrophic forgetting). UFO claims you can inject rare motions (cartwheel) without collapsing learned behavior. If it holds, that's a significant incremental/continual skill-learning! But again, I have to underline it: no arXiv, no external validation, no success-rate numbers. -> robotics badely needs an independent unbiased evaluator imho. Still, look at that cool demo: robot is getting kicked and pushed around (serious disturbance) during teleoperation (controlled the person at the back wearing the VR headset), and still managed to always get back up. This is some serious demonstration of stability and robustness!
0:49

New framework: Kick down your robot, it will get back up every time 🥋 Chinese startup RoboParty is a Beijing startup founded April 2025 by Huang Yi, originally shipping ROBOTO ORIGIN, the world's first full-stack open-source bipedal humanoid. They released UFO: Unsupervised Reinforcement Learning Framework for Humanoid Control. DEFINITIONS -> what differs is where the learning signal comes from: - SUPERVISED: humans supply the right answers (labels), the model imitates them. - UNSUPERVISED: no answer key, the model finds structure in raw data on its own. - REINFORCEMENT LEARNING: no answer key either, the model tries things and a reward scores each attempt. → UNSUPERVISED RL: trial and error where the agent invents its own rewards, instead of engineers hand-writing one per task. REPRESENTATION LEARNING: compress raw states into a useful internal map. TEMPORAL DISTANCE: distance on that map is "how many steps from A to B." CONTRASTIVE: trained by pulling together what's close in time, pushing apart what isn't. -> CONTRASTIVE TEMPORAL-DISTANCE REPRESENTATION LEARNING: the model builds an internal map of body states where distance means how many steps it takes to get from one to another. It is trained by contrast: states that occur close together in a movement get pulled together in the map, randomly paired states get pushed apart. UFO is an open-source training framework that teaches humanoid robots skills, like getting up, walking, goal-reaching, teleoperation, without reference motions -> no motion-capture or human-video demonstrations to imitate. Its core is TeCH, a contrastive temporal-distance representation-learning algorithm: the robot explores, builds pseudo-goals by temporal rolling, and learns goal-conditioned policies from a single unified progress reward. One framework trains five different robots (Unitree G1/H1, RoboParty RP0/RP1, AgiBot X2) with automatic config conversion in ~2–3 hours per robot! The real novelty here "no demonstrations at all". No data-collection arms race,the dominant humanoid-locomotion recipe is tracking: imitate mocap/retargeted-human reference trajectories. The robot self-generates goals from its own exploration and learns from a progress reward, needing zero reference motion data. Everybody else is fighting over data acquisition, while this team just teleports out of the race entirely (inb4 "competition is for losers 💀 ). This strategy reminds me of the DeepSeek playbook applied to robots: open-source the whole stack to become the global default and commoditize everyone else. RoboParty is giving away hardware and now control software (UFO) to be the Android of humanoids. Yet another reason for the US to ban Chinese open models perhaps 🥶 ? What I also really like about this approach is the cross-embodiment infrastructure, one framework trains Unitree G1/H1, RoboParty RP0/RP1, and AgiBot X2 with automatic configuration conversion. Just like Physical Intelligence, RoboParty seems to place itself as a neutral hardware agnostic middle man. Also woth mentioning: their ability ot perform stable skill injection, e.g. adding a cartwheel without forgetting how to walk. A common failure of RL humanoid policies is that teaching a new agile skill destabilizes the existing ones (catastrophic forgetting). UFO claims you can inject rare motions (cartwheel) without collapsing learned behavior. If it holds, that's a significant incremental/continual skill-learning! But again, I have to underline it: no arXiv, no external validation, no success-rate numbers. -> robotics badely needs an independent unbiased evaluator imho. Still, look at that cool demo: robot is getting kicked and pushed around (serious disturbance) during teleoperation (controlled the person at the back wearing the VR headset), and still managed to always get back up. This is some serious demonstration of stability and robustness!

Léo

36,184 просмотров • 2 месяцев назад

Today we’re launching the first and only human-like AI agents in the world. Super Agents™ are the first agents with human‑level skills – they DM you, take @ mentions, send emails, manage docs, tasks, and more. Not just tools or API calls, but real skills fine‑tuned for how teams actually work. The first agents with 100% context – fully native in ClickUp and fully synced from other apps. Super Agents see your work the same way that humans do: tasks, docs, schedules, and conversations all in one place. The first agents that learn from human interactions automatically, without any setup or configuration – when you give feedback, they listen and improve how they work. The first agents with human‑level memory for custom agents – historical memory for every interaction, short-term working memory, and even long‑term memory stored in docs you can literally open, inspect, and edit. The first agents that are literally the same as users – our agentic user model is the same as our user data model. This gives you permissions and capabilities that you and your systems are already familiar with. The first infinite agent catalog – where anyone can create and customize agents in minutes, for literally any type of work imaginable. It's the most intuitive way to build agents on the planet. 95% of companies are failing in AI adoption. The reality is that AI isn't meant to be adopted, it's meant to be adapted – to you. Super Agents are automatically personalized to you and your company using proprietary state-of-the-art agent architecture, orchestration, and tooling. Today is the largest step forward we've ever made towards our mission of making people more productive. Maximize human productivity, with ClickUp Super Agents. Available NOW. For everyone.
5:50

Today we’re launching the first and only human-like AI agents in the world. Super Agents™ are the first agents with human‑level skills – they DM you, take @ mentions, send emails, manage docs, tasks, and more. Not just tools or API calls, but real skills fine‑tuned for how teams actually work. The first agents with 100% context – fully native in ClickUp and fully synced from other apps. Super Agents see your work the same way that humans do: tasks, docs, schedules, and conversations all in one place. The first agents that learn from human interactions automatically, without any setup or configuration – when you give feedback, they listen and improve how they work. The first agents with human‑level memory for custom agents – historical memory for every interaction, short-term working memory, and even long‑term memory stored in docs you can literally open, inspect, and edit. The first agents that are literally the same as users – our agentic user model is the same as our user data model. This gives you permissions and capabilities that you and your systems are already familiar with. The first infinite agent catalog – where anyone can create and customize agents in minutes, for literally any type of work imaginable. It's the most intuitive way to build agents on the planet. 95% of companies are failing in AI adoption. The reality is that AI isn't meant to be adopted, it's meant to be adapted – to you. Super Agents are automatically personalized to you and your company using proprietary state-of-the-art agent architecture, orchestration, and tooling. Today is the largest step forward we've ever made towards our mission of making people more productive. Maximize human productivity, with ClickUp Super Agents. Available NOW. For everyone.

Zeb Evans

320,989 просмотров • 9 месяцев назад