正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Introducing 𝐀𝐋𝐎𝐇𝐀 𝐔𝐧𝐥𝐞𝐚𝐬𝐡𝐞𝐝 🌋 - Pushing the boundaries of dexterity with low-cost robots and AI. Google DeepMind Finally got to share some videos after a few months. Robots are fully autonomous filmed in one continuous shot. Enjoy!

Tony Zhao

126,360 subscribers

354,259 次观看 • 2 年前 •via X (Twitter)

科学技术艺术

Anya Rossi• Live Now

Private livecam show

10 条评论

Tony Z. Zhao 的头像

Tony Z. Zhao2 年前

Check out the tweet from @ayzwah for more details and closeup videos!

Tony Z. Zhao 的头像

Tony Z. Zhao2 年前

team: @ayzwah, @JonathanTompson, @DannyDriess, @peteflorence, @coolboi95, @chelseabfinn @SpencerGoodric6 🚀

Nat Friedman 的头像

Nat Friedman2 年前

@GoogleDeepMind Wow!

cole murray 的头像

cole murray2 年前

@GoogleDeepMind The second hand stabilizing 🔥👏

Peter Chen 的头像

Peter Chen2 年前

@GoogleDeepMind Love the one continuous take demo! Congrats @tonyzzhao !

Investronauts 的头像

Investronauts2 年前

@GoogleDeepMind Is the model open-source?

sarah guo // conviction 的头像

sarah guo // conviction2 年前

@GoogleDeepMind GOATED the robots are coming

Gonzalo Espinoza Graham 🏴‍☠️ 的头像

Gonzalo Espinoza Graham 🏴‍☠️2 年前

@GoogleDeepMind This demo is like traveling in time and seeing what the future holds for us. Exciting

Andre Infante 的头像

Andre Infante2 年前

@GoogleDeepMind This is really nice work! How much data are we looking at here? Is it one policy with task conditioning or a bunch of separate checkpoints?

Seif 的头像

Seif2 年前

@GoogleDeepMind really cool! how many demonstrations did you need on average per task?

相关视频

Introducing the world's smallest, fully programmable, autonomous robots. Created by Marc Miskin of @Penn_ESE and David Blauuw of Michigan Engineering, the microscopic robots can independently sense and respond to their surroundings, operate for months and cost just a penny each.

Introducing the world's smallest, fully programmable, autonomous robots. Created by Marc Miskin of @Penn_ESE and David Blauuw of Michigan Engineering, the microscopic robots can independently sense and respond to their surroundings, operate for months and cost just a penny each.

Penn Engineering

18,213 次观看 • 8 个月前

From cleaning up the kitchen to fetching a snack, could robots one day assist us in our daily lives? In Google DeepMind: The Podcast, Jeff Dean spoke with Hannah Fry about the potential for AI to power robots capable of understanding and responding to natural language. ↓

From cleaning up the kitchen to fetching a snack, could robots one day assist us in our daily lives? In Google DeepMind: The Podcast, Jeff Dean spoke with Hannah Fry about the potential for AI to power robots capable of understanding and responding to natural language. ↓

Google DeepMind

126,197 次观看 • 1 年前

Robots are not ready to take over the world yet! Zipeng Fu and I just compiled a video of the dumbest mistakes 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐋𝐎𝐇𝐀🏄 made in the autonomous mode 🤣 We are also planning to organize some live demos after taking a break. Stay tuned!

Robots are not ready to take over the world yet! Zipeng Fu and I just compiled a video of the dumbest mistakes 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐋𝐎𝐇𝐀🏄 made in the autonomous mode 🤣 We are also planning to organize some live demos after taking a break. Stay tuned!

Tony Zhao

332,809 次观看 • 2 年前

Boston Dynamics and Google DeepMind announced a partnership to integrate Gemini Robotics AI models with the new electric Atlas humanoid robots. Got to say hi to Carolina Parada before the presentation

Boston Dynamics and Google DeepMind announced a partnership to integrate Gemini Robotics AI models with the new electric Atlas humanoid robots. Got to say hi to Carolina Parada before the presentation

The Humanoid Hub

14,689 次观看 • 7 个月前

Boston Dynamics will begin production of the new Atlas robots at its Boston headquarters immediately. All Atlas deployments are already fully committed for 2026, with fleets scheduled to ship to Hyundai and Google DeepMind in the coming months. The company plans to add additional customers in early 2027 and also plans to build a new robotics factory by 2028 that is capable of producing 30,000 robots per year.

Boston Dynamics will begin production of the new Atlas robots at its Boston headquarters immediately. All Atlas deployments are already fully committed for 2026, with fleets scheduled to ship to Hyundai and Google DeepMind in the coming months. The company plans to add additional customers in early 2027 and also plans to build a new robotics factory by 2028 that is capable of producing 30,000 robots per year.

The Humanoid Hub

349,617 次观看 • 7 个月前

Time to democratize humanoid robots! Introducing ToddlerBot, a low-cost ($6K), open-source humanoid for robotics and AI research. Watch two ToddlerBots seamlessly chain their loco-manipulation skills to collaborate in tidying up after a toy session.

Time to democratize humanoid robots! Introducing ToddlerBot, a low-cost ($6K), open-source humanoid for robotics and AI research. Watch two ToddlerBots seamlessly chain their loco-manipulation skills to collaborate in tidying up after a toy session.

Haochen Shi

113,527 次观看 • 1 年前

For the past year we've been working on ALOHA Unleashed 🌋 @GoogleDeepmind - pushing the scale and dexterity of tasks on our ALOHA 2 fleet. Here is a thread with some of the coolest videos! The first task is hanging a shirt on a hanger (autonomous 1x)

For the past year we've been working on ALOHA Unleashed 🌋 @GoogleDeepmind - pushing the scale and dexterity of tasks on our ALOHA 2 fleet. Here is a thread with some of the coolest videos! The first task is hanging a shirt on a hanger (autonomous 1x)

Ayzaan Wahid

141,032 次观看 • 2 年前

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 次观看 • 3 年前

How can robots acquire fine-grained manipulation skills? Introducing ACT: Action Chunking with Transformers 🤖 Key idea: Imitation, but predict actions in chunks instead of one at a time. Here are results with only ~15min of demonstrations, running on low-cost arms:

How can robots acquire fine-grained manipulation skills? Introducing ACT: Action Chunking with Transformers 🤖 Key idea: Imitation, but predict actions in chunks instead of one at a time. Here are results with only ~15min of demonstrations, running on low-cost arms:

Tony Zhao

247,660 次观看 • 3 年前

🇯🇵 1 in 3 Japanese companies are already using or thinking about deploying AI-powered robots, with automakers leading the charge. The driver is simple: Japan is running out of workers and robots are the plan. Japan just needs to hold off China and the U.S. first. They are giving Japan serious competition in the next generation of autonomous robots, and its old industrial giants don't have the edge they once did. Source: Reuters

🇯🇵 1 in 3 Japanese companies are already using or thinking about deploying AI-powered robots, with automakers leading the charge. The driver is simple: Japan is running out of workers and robots are the plan. Japan just needs to hold off China and the U.S. first. They are giving Japan serious competition in the next generation of autonomous robots, and its old industrial giants don't have the edge they once did. Source: Reuters

Mario Nawfal

27,617 次观看 • 2 个月前

Dynamic manipulation steps up to the plate! This is a first look at a low-impedance platform designed to study how robots manipulate objects. In this demo, two robots play catch and practice batting, even teaming up with humans. The robots are capable of throwing 70mph [112 kph] and can catch and bat at short distances (23ft [7m]) :

Dynamic manipulation steps up to the plate! This is a first look at a low-impedance platform designed to study how robots manipulate objects. In this demo, two robots play catch and practice batting, even teaming up with humans. The robots are capable of throwing 70mph [112 kph] and can catch and bat at short distances (23ft [7m]) :

RAI Institute

25,497 次观看 • 9 个月前

XPENG founder and CEO He Xiaopeng announced that they will release a new generation of humanoid robots next month, introducing a new AI system, VLT, combined with five other systems, including VLA/VLM. The goal is to transition humanoid robots from remote-controlled to fully autonomous. Their second-generation humanoid robot, Iron, released last year, is already undergoing trial operations at the XPENG factory in Guangzhou and serving as a shopping guide in the flagship store. It's worth noting that Tesla plans to release the Optimus V3 by the end of the year.

XPENG founder and CEO He Xiaopeng announced that they will release a new generation of humanoid robots next month, introducing a new AI system, VLT, combined with five other systems, including VLA/VLM. The goal is to transition humanoid robots from remote-controlled to fully autonomous. Their second-generation humanoid robot, Iron, released last year, is already undergoing trial operations at the XPENG factory in Guangzhou and serving as a shopping guide in the flagship store. It's worth noting that Tesla plans to release the Optimus V3 by the end of the year.

CyberRobo

12,948 次观看 • 10 个月前

Introducing 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐋𝐎𝐇𝐀🏄 -- Hardware! A low-cost, open-source, mobile manipulator. One of the most high-effort projects in my past 5yrs! Not possible without co-lead Zipeng Fu and Chelsea Finn. At the end, what's better than cooking yourself a meal with the 🤖🧑‍🍳

Introducing 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐋𝐎𝐇𝐀🏄 -- Hardware! A low-cost, open-source, mobile manipulator. One of the most high-effort projects in my past 5yrs! Not possible without co-lead Zipeng Fu and Chelsea Finn. At the end, what's better than cooking yourself a meal with the 🤖🧑‍🍳

Tony Zhao

1,668,615 次观看 • 2 年前

ELON: AI AND ROBOTS WILL CREATE AN AGE OF ABUNDANCE "I think that there's no actual limit to the size of the economy once you have general purpose humanoid robots and autonomous vehicles. You can build anything. The economy is really just the average productivity per person times number of people. That's the economy, and if you've got humanoid robots where there's no real limit on the number of humanoid robots and they can operate very intelligently, then there's no meaningful limit to the economy. The good future of AI is one of immense prosperity, where there is an age of abundance, no shortage of goods and services. Everyone can have whatever they want, anything that is a manufactured good or provided service. With the advent of AI plus robotics, the cost of goods and services will trend to zero." Source: The All-In Podcast

ELON: AI AND ROBOTS WILL CREATE AN AGE OF ABUNDANCE "I think that there's no actual limit to the size of the economy once you have general purpose humanoid robots and autonomous vehicles. You can build anything. The economy is really just the average productivity per person times number of people. That's the economy, and if you've got humanoid robots where there's no real limit on the number of humanoid robots and they can operate very intelligently, then there's no meaningful limit to the economy. The good future of AI is one of immense prosperity, where there is an age of abundance, no shortage of goods and services. Everyone can have whatever they want, anything that is a manufactured good or provided service. With the advent of AI plus robotics, the cost of goods and services will trend to zero." Source: The All-In Podcast

Mario Nawfal

341,719 次观看 • 9 个月前

Time for the open-source AI robots revolution 🚀 We’ve been playing with a low-cost DJI robot controlled by 3 local open-source AI models (Whisper, Idefics2, Parler-TTS - all Apache2) & orchestrated by Dora-cs In comments a 250 lines code gist to build on top of it => enjoy!!

Time for the open-source AI robots revolution 🚀 We’ve been playing with a low-cost DJI robot controlled by 3 local open-source AI models (Whisper, Idefics2, Parler-TTS - all Apache2) & orchestrated by Dora-cs In comments a 250 lines code gist to build on top of it => enjoy!!

Thomas Wolf

138,089 次观看 • 2 年前

.Sergey Levine is one of the world's leading robotics researchers (and co-founder of Physical Intelligence). He thinks fully autonomous robots are much closer than people realize - when I pushed him on a prediction, he said 5 years to robots that can autonomously run your household). The big game changer, in his view, is that we finally have LLMs with common sense and prior knowledge that we can scaffold the robot models around. We get into all the reasons this might not happen, the data flywheel from deployment, architecture, simulation, manufacturing, and much more. Enjoy! 0:00:00 – Timeline to widely deployed autonomous robots 0:22:12 – Why robotics will scale faster than self-driving cars 0:32:15 – How vision-language-action models work 0:50:26 – Improvements needed for brainlike efficiency 1:02:48 – Learning from simulation 1:14:08 – How much will robots speed up AI buildouts? 1:22:54 – If hardware’s the bottleneck, does China win by default?

.Sergey Levine is one of the world's leading robotics researchers (and co-founder of Physical Intelligence). He thinks fully autonomous robots are much closer than people realize - when I pushed him on a prediction, he said 5 years to robots that can autonomously run your household). The big game changer, in his view, is that we finally have LLMs with common sense and prior knowledge that we can scaffold the robot models around. We get into all the reasons this might not happen, the data flywheel from deployment, architecture, simulation, manufacturing, and much more. Enjoy! 0:00:00 – Timeline to widely deployed autonomous robots 0:22:12 – Why robotics will scale faster than self-driving cars 0:32:15 – How vision-language-action models work 0:50:26 – Improvements needed for brainlike efficiency 1:02:48 – Learning from simulation 1:14:08 – How much will robots speed up AI buildouts? 1:22:54 – If hardware’s the bottleneck, does China win by default?

Dwarkesh Patel

211,158 次观看 • 11 个月前

Running legged robots with AI. Today we met @swiss_mile, the team connecting AI with the physical world using autonomous machines. An ideal mobile robot needs to move quickly, overcoming obstacles while carrying heavy loads—something current robots still struggle with. Many robots haven't been widely adopted due to issues with real-world problem-solving. To address this, the Swiss-mile team have created a solution that combines robots with wheels and legs, powered by advanced AI. These robots can walk, drive, stand, open doors, and even handle packages. By integrating these features, the team are on a mission to push automation forward beyond stationary robotic arms and have raised $22m in Seed Funding from Jeff Bezos, HongShan, Amazon Industrial Innovation Fund and Armada Investment to do so.

Running legged robots with AI. Today we met @swiss_mile, the team connecting AI with the physical world using autonomous machines. An ideal mobile robot needs to move quickly, overcoming obstacles while carrying heavy loads—something current robots still struggle with. Many robots haven't been widely adopted due to issues with real-world problem-solving. To address this, the Swiss-mile team have created a solution that combines robots with wheels and legs, powered by advanced AI. These robots can walk, drive, stand, open doors, and even handle packages. By integrating these features, the team are on a mission to push automation forward beyond stationary robotic arms and have raised $22m in Seed Funding from Jeff Bezos, HongShan, Amazon Industrial Innovation Fund and Armada Investment to do so.

Circuit

17,876 次观看 • 1 年前