Chris Paxton's banner

Chris Paxton

@chris_j_paxton • 34,315 subscribers

Mostly posting about robots. currently AI @agilityrobotics prev embodied AI @AIatMeta, @NVIDIAAI. All views my own.

Shorts

Always important to remember that a lot of these robots are "faking" the humanlike motions -- its a property of how they're trained not an inherent property of the hardware. They're actually capable of way weirder stuff and way faster motions.

Always important to remember that a lot of these robots are "faking" the humanlike motions -- its a property of how they're trained not an inherent property of the hardware. They're actually capable of way weirder stuff and way faster motions.

9,866,635 次观看

This is looking amazing

This is looking amazing

254,823 次观看

Berry pickup from Kyber Labs Robot hands that can pick up fragile objects quickly will be important to solving long-tail of robotics problems not currently automated

Berry pickup from Kyber Labs Robot hands that can pick up fragile objects quickly will be important to solving long-tail of robotics problems not currently automated

33,336 次观看

Predicting how the environment changes over time is a key task in robotics, enabling all kinds of downstream tasks. Trace Anything is a very interesting look at how you can train models which predict the trajectory each point will take in the future -- you can use this for robot manipulation, video prediction, and more. Learn more ->

Predicting how the environment changes over time is a key task in robotics, enabling all kinds of downstream tasks. Trace Anything is a very interesting look at how you can train models which predict the trajectory each point will take in the future -- you can use this for robot manipulation, video prediction, and more. Learn more ->

56,845 次观看

Watching the agibot x2 at the hackathon yesterday. I think there are a few extra degrees of freedom over the unitree g1 and you can really see it in the motion

Watching the agibot x2 at the hackathon yesterday. I think there are a few extra degrees of freedom over the unitree g1 and you can really see it in the motion

22,281 次观看

Pantograph robot building with jenga blocks. Focusing on RL is a great and fairly unique strategy

Pantograph robot building with jenga blocks. Focusing on RL is a great and fairly unique strategy

30,546 次观看

A favorite of the new robot designs from Fourier -- the gr3 is soft(ish) and friendly

A favorite of the new robot designs from Fourier -- the gr3 is soft(ish) and friendly

44,042 次观看

These crazy videos of Chinese delivery robots from Neolix (and maybe similar companies) are my new favorite thing

These crazy videos of Chinese delivery robots from Neolix (and maybe similar companies) are my new favorite thing

19,236 次观看

Some random thoughts reading through the new RobbyAnt VLA paper: - 20,000 hours of data across 9 robots!! - damn, Chinese companies are going to trivially outscale the American ones on real robot data - It's really cool they train on depth; it means they can handle transparent objects really well for example - You can never tell how good these models are without trying them, since everyone trains on different robots, but from the results they show it does seem to clean up - cross embodiment scaling laws are really cool - if you have lots of robots do you need human video??

Some random thoughts reading through the new RobbyAnt VLA paper: - 20,000 hours of data across 9 robots!! - damn, Chinese companies are going to trivially outscale the American ones on real robot data - It's really cool they train on depth; it means they can handle transparent objects really well for example - You can never tell how good these models are without trying them, since everyone trains on different robots, but from the results they show it does seem to clean up - cross embodiment scaling laws are really cool - if you have lots of robots do you need human video??

19,921 次观看

Learn about how to make humanoid robot teleoperation work for humanoid *mobile* manipulation in this podcast. Really cool work by Siyuan Huang and yixuan_li!

Learn about how to make humanoid robot teleoperation work for humanoid mobile manipulation in this podcast. Really cool work by Siyuan Huang and yixuan_li!

25,060 次观看

ManiFeel is a benchmark for contact-rich manipulation tasks. Force- and tactile-based tasks with occlusions and partially-observable state are going to be crucial for true general purpose robotics. Website:

ManiFeel is a benchmark for contact-rich manipulation tasks. Force- and tactile-based tasks with occlusions and partially-observable state are going to be crucial for true general purpose robotics. Website:

12,063 次观看

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Wuji hands used by genesis to do a wire harnessing task. This kind of thing was completely out of sight of robotics until a couple years ago and is now possible with off the shelf hardware.

Wuji hands used by genesis to do a wire harnessing task. This kind of thing was completely out of sight of robotics until a couple years ago and is now possible with off the shelf hardware.

109,494 次观看 • 6 天前

Reach the banana hanging from the ceiling

Reach the banana hanging from the ceiling

69,113 次观看 • 1 个月前

The best humanoid robotics in the world from 10 years ago at the Darpa grand challenge. Its amazing how fast this has changed; imagine what this will be like in 10 years

The best humanoid robotics in the world from 10 years ago at the Darpa grand challenge. Its amazing how fast this has changed; imagine what this will be like in 10 years

291,368 次观看 • 7 个月前

1x neo gamma vacuuming at GTC

1x neo gamma vacuuming at GTC

515,783 次观看 • 1 年前

Cool look at a fleet of Apollo 2 robots doing a variety of pick and place tasks. Actually this whole video showed quite a wide range of uses for their robot.

Cool look at a fleet of Apollo 2 robots doing a variety of pick and place tasks. Actually this whole video showed quite a wide range of uses for their robot.

25,686 次观看 • 22 天前

Easily the most impressive uncut autonomous video I've ever seen, from Physical Intelligence

Easily the most impressive uncut autonomous video I've ever seen, from Physical Intelligence

467,648 次观看 • 1 年前

This is the first time that I have seen Skild doing live demos in public! Demonstrating neural nets for precision manufacturing

This is the first time that I have seen Skild doing live demos in public! Demonstrating neural nets for precision manufacturing

107,554 次观看 • 4 个月前

The best way to get robust, high-quality robot performance is through reinforcement learning; but RL in either the real world or a traditional simulation has lots of limitations. Instead, Jiazhi Yang in RISE does RL in a compositional world model. Learn more ->

The best way to get robust, high-quality robot performance is through reinforcement learning; but RL in either the real world or a traditional simulation has lots of limitations. Instead, Jiazhi Yang in RISE does RL in a compositional world model. Learn more ->

33,766 次观看 • 1 个月前

It is really incredible what these robots can do now from omnixtreme - new RoboPapers episode tomorrow

It is really incredible what these robots can do now from omnixtreme - new RoboPapers episode tomorrow

62,336 次观看 • 2 个月前

Gears are the "muscles" of robotics; but with so many AI/ML folks getting into robotics now, you may not know what people mean when they say cycloid/strain wave/planetary gear, and, more importantly why they matter What are the advantages and disadvantages? What actually ARE these things? I made this over the weekend to try to help provide an intuition: Probably has some mistakes as I am no mechanical engineer but I think it's useful

Gears are the "muscles" of robotics; but with so many AI/ML folks getting into robotics now, you may not know what people mean when they say cycloid/strain wave/planetary gear, and, more importantly why they matter What are the advantages and disadvantages? What actually ARE these things? I made this over the weekend to try to help provide an intuition: Probably has some mistakes as I am no mechanical engineer but I think it's useful

16,526 次观看 • 22 天前

The kind of things you can do with compliant/force aware teleop i guess. Another cool video of a highly constrained pick.

The kind of things you can do with compliant/force aware teleop i guess. Another cool video of a highly constrained pick.

21,481 次观看 • 1 个月前

Still one of my favorite robotics use cases: Peanut Robotics cleaning hotels and workplaces autonomously. Need a home version of this guy

Still one of my favorite robotics use cases: Peanut Robotics cleaning hotels and workplaces autonomously. Need a home version of this guy

236,073 次观看 • 2 年前

Its funny how even with the wonderfully dynamic unitree g1, these actions are slow. Perception-> action and the embodiment gap are the big things slowing down robotics. This is the part of the video where they start manipulating stuff

Its funny how even with the wonderfully dynamic unitree g1, these actions are slow. Perception-> action and the embodiment gap are the big things slowing down robotics. This is the part of the video where they start manipulating stuff

94,483 次观看 • 8 个月前

UMI-style robot learning with mobile manipulation and multiple robots

UMI-style robot learning with mobile manipulation and multiple robots

12,148 次观看 • 23 天前

Its cool to see this robot stand up

Its cool to see this robot stand up

47,134 次观看 • 4 个月前

This is what I want to see from a hand video, damn

This is what I want to see from a hand video, damn

50,230 次观看 • 5 个月前

One of my favorite works in the "robot learning from human videos" space that I have seen lately. This is because it offers a really coherent and unique way to represent robot skills without a lot of the downsides we see from estimation-based approaches.

One of my favorite works in the "robot learning from human videos" space that I have seen lately. This is because it offers a really coherent and unique way to represent robot skills without a lot of the downsides we see from estimation-based approaches.

78,419 次观看 • 9 个月前

Still one of the top "oh wait this might really work" kind of moments for me

Still one of the top "oh wait this might really work" kind of moments for me

46,687 次观看 • 5 个月前

The future of robot butlers starts with mobile manipulation. We’re announcing the NeurIPS 2023 Open-Vocabulary Mobile Manipulation Challenge! - Full robot stack ✅ - Parallel sim and real evaluation ✅ - No robot required ✅👀

The future of robot butlers starts with mobile manipulation. We’re announcing the NeurIPS 2023 Open-Vocabulary Mobile Manipulation Challenge! - Full robot stack ✅ - Parallel sim and real evaluation ✅ - No robot required ✅👀

178,966 次观看 • 3 年前

This clip from Figure shows what's so exciting about end to end applications. Handling a messy, unstructured feed of objects like this would be very difficult to scale with traditional methods

This clip from Figure shows what's so exciting about end to end applications. Handling a messy, unstructured feed of objects like this would be very difficult to scale with traditional methods

79,868 次观看 • 1 年前