🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

We will present HACMan++ at #RSS2024 at 8:30 am tomorrow (Friday). Come after to our poster to chat about our new contact-centric action space for reinforcement learning!

We are looking for great demos at CoRL conference in Munich. Submit by August 2nd to get early feedback. The final proposal at August 16th 2 pages. Details of the call: corl.org/contributions/… Please RT.

#RSS2024 was so inspiring - both the robots and the random objects I encountered!

Maybe there is a sweet spot in between which combines System-2 + System-1 reasoning like what we proposed in LOOP (MPC with a terminal value) (blog.ml.cmu.edu/2022/01/07/loo…) and nicklashansen proposes in TD-MPC (improves upon similar principles adding a latent dynamics model)

Anyone have slides on Canny edge detection that don't use Lena that I can use? (I will give you credit in my slides). Thanks in advance!

One thing that is VERY hard to get across in a video is the concept of generalization. You can make a beautiful robot demo if you're testing in a very similar environment to what you trained in. If we want robots to work in our homes, though, they need to perform complex tasks on

How can an autonomous agent leverage novel tools to cut, roll, and scoop a piece of dough, given just a few tool shapes for training? Our method generates a “desired tool shape” that performs the motion and then matches the real tool to the generated shape.sites.google.com/view/toolgen

I am excited to share our latest work on teaching robots to use novel tools! Our results show that intelligent tool-use skills can be learned through a general framework of predicting task-conditioned pointcloud trajectories. Check out the detailed thread by Carl Qi.

1/ 🎉 How can we develop methods to generate synthetic, photorealistic data for training #AI models in robotics? We present SplatSim, a step in this direction. SplatSim is a scalable framework that generates photorealistic data for manipulation tasks using existing simulators as

Introducing TAX3D, in which we extend relative-placement methods to generalizable deformable manipulation! #CoRL2024 (1/🧵) Our approach generalizes to: - Diverse unseen objects - Diverse unseen configurations - Multimodal placements

Excited to share our recent work on CoRL2024, We enable quadruped robots to manipulate (flip &push) objects combined with locomotion. The whole visual policy is trained end2end by RL and controlled by MPC. More details can be found on our website: legged-manipulation.github.io

Checkout our new NeurIPS paper on using differentiable trajectory optimization for deep RL and IL! Join our poster session at Friday 11-2. Ziyu will present this paper (he is also applying for grad schools this cycle)

Accepted to ICRA, 2025!🎉

I am Jewish and a faculty. When the current administration cancels grants and attempts to deport students in the name of "protecting Jews from antisemitism" they are not speaking for me or defending me. They are using us as a totem to attack universities and immigration

A colleague mentioned this to me today and I just checked it: ChatGPT still cannot count the number of r's in strrrawberrrrry: chatgpt.com/share/67dd8bef…

Does anyone have a copy of the ShapeNetSem dataset? The Sshapenet website is having issues for downloading the dataset. Thanks!!

Check out our latest work on opening unseen articulated objects in the real world; this is a single policy that operates zero-shot on 3 robot embodiments! Key insights: 1) Large-scale sim training 2) Hierarchical imitation learning