Ville🤖's banner

Ville🤖

@VilleKuosmanen • 3,519 subscribers

gentleman scientist 🤖 @armnet_dev building the world's largest open arm farm 🇫🇮🇬🇧

Shorts

Benchmarking robot policies is broken We ran thousands of evaluations over 12 different manipulation tasks to evaluate which robot policy you should use.

Benchmarking robot policies is broken We ran thousands of evaluations over 12 different manipulation tasks to evaluate which robot policy you should use.

28,344 views

I rented an H100 and fine-tuned Physical Intelligence ‘s pi0 overnight. We’re still far from fully utilising the potential of this model, but it produced my first successful rollout!

I rented an H100 and fine-tuned Physical Intelligence ‘s pi0 overnight. We’re still far from fully utilising the potential of this model, but it produced my first successful rollout!

66,403 views

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

First demo of reproducing force feedback from the FACTR 2 paper It relies on torque and current sensing which the feetechs don’t handle as well due to large gear ratios so it’s not perfect. Not sure if limits are in hw or we can do better with core changes & more data

First demo of reproducing force feedback from the FACTR 2 paper It relies on torque and current sensing which the feetechs don’t handle as well due to large gear ratios so it’s not perfect. Not sure if limits are in hw or we can do better with core changes & more data

19,216 views • 1 month ago

A little weekend side-project, trying out gravity comp with PWM controls on an so-101 It holds but feels very stiff, the giant gear ratios probably impact backdriveability. Will see if any AI/SW tricks can mitigate, would prefer keeping standard servos Pepijn 🦾

A little weekend side-project, trying out gravity comp with PWM controls on an so-101 It holds but feels very stiff, the giant gear ratios probably impact backdriveability. Will see if any AI/SW tricks can mitigate, would prefer keeping standard servos Pepijn 🦾

16,106 views • 2 months ago

Do AI robots see the world like we do? I dove head first into latent space to uncover the attention maps that show how my robot sees and understands the world.

Do AI robots see the world like we do? I dove head first into latent space to uncover the attention maps that show how my robot sees and understands the world.

44,266 views • 1 year ago

MolmoPoint from Ai2 was a recent foundation model release that feels criminally underrated to me, while not a robotics model it has great potential for various use cases labels are great even for wrist view!

MolmoPoint from Ai2 was a recent foundation model release that feels criminally underrated to me, while not a robotics model it has great potential for various use cases labels are great even for wrist view!

20,015 views • 6 months ago

RewACT, my reward estimation mod for ACT policies, is now available on PyPi, making it significantly easier to install and use!

RewACT, my reward estimation mod for ACT policies, is now available on PyPi, making it significantly easier to install and use!

29,935 views • 1 year ago

Another uncut video of robots manipulating objects - this time it only missed a grasp at the very end. One step closer towards a general method of optimising robot reliability - still a long way to go though!

Another uncut video of robots manipulating objects - this time it only missed a grasp at the very end. One step closer towards a general method of optimising robot reliability - still a long way to go though!

24,569 views • 1 year ago

Going back to training single task models with a focus on improving success rates - here’s a longer, unedited real-time rollout of an ACT model trained with 40 episodes. Explanation in 🧵

Going back to training single task models with a focus on improving success rates - here’s a longer, unedited real-time rollout of an ACT model trained with 40 episodes. Explanation in 🧵

11,845 views • 1 year ago

No more content to load