Profile avatar
traversaro.bsky.social
Robotics Applied Scientist @ami_iit @iCub @IITalk . Robotics Software @condaforge @RoboStack .
15 posts 341 followers 826 following
Regular Contributor
Conversation Starter

🚀 Great news for Windows users! The latest version of PyTorch is back! 🥳 🙏 Huge thanks to the @condaforge community for fixing the builds. 🧑‍💻 Installing PyTorch just got a whole lot easier! pixi.sh/latest/feat...

Introducing playground.mujoco.org Combining MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms. Get started today: pip install playground

Install a GPU enabled PyTorch on your machine (with all CUDA dependencies if on Linux) from @conda-forge.org in two commands with @prefix.dev's pixi: github.com/matthewfeick...

Robot end-to-end learning gets better when including force medium.com/correll-l...

Remember: it is a calmer more chill life without copyright years. https://daniel.haxx.se/blog/2023/01/08/copyright-without-years/

2024 was a busy one! We released v 1.0 of the Open Source AI Definition, among many other accomplishments across 3 grand areas: licensing and legal, policy and standards, and research and outreach: https://opensource.org/2024-annual-report

🎉 Our new work tackling long horizon low-level manipulation in apartments is out! ~500GB of demonstration data in sim (you can generate more) and RL/IL baselines all provided. ManiSkill helped make this project scalable and faster to run via GPU sim compared to alternatives

📢 Introducing ManiSkill-HAB: A benchmark for low-level manipulation in home rearrangement tasks! - GPU-accelerated simulation - Extensive RL/IL baselines - Vision-based, whole-body control robot dataset All open-sourced: arth-shukla.github.io/mshab 🧵(1/5)

popular computer vision package ultralytics (home of yolov8 and yolo11) was compromised. a crypto miner was injected into versions 8.3.41 and 8.3.42. link: github.com/ultralytics/...

Check out Silvio Traversaro's amazing presentation on using Pixi in robotics research! Watch the recording here: www.youtube.com/live/fvgfTv... #Robotics #Pixi #Research #TechTalk

Check out Pablo Vela Gomez's presentation at our Show & Tells! 🎥 Discover the ultimate 3D vision development experience with pixi, @rerundotio, @huggingface, and @Gradio. Watch now: www.youtube.com/watchv=ff7Bh... This video is one of the examples in his presentation:

Happy to announce the PySR v1.0 release! github.com/MilesCranmer... PySR lets you do high-performance symbolic regression from Python. Now, you can learn multiple symbolic expressions simultaneously! Also: + Parametric expressions + TensorBoard support + Improved search + Julia-based inference

We are looking for the current best multi-view full-body 3d pose estimation model/software with Remi Cadene Any good advice? Should include hands pose estimation in addition to body preferably Better if able to use multiple cameras as inputs (multi-view) for open-source low cost robot teleop

(Other Twitter stuff) Human-robot and robot-robot collaborative carrying with mobile robots typically requires a manipulator arm. At iros2024, we introduce PACC, a novel passive-arm design that, coupled with a decentralized MPC, allows an efficient execution of such tasks with quadruped robots.

There surely must be more people here working on open source robotics? Please let me know who else to add.

(Posting old Twitter stuff for boosting connections/the algorithm) Open-source (and in python!) MPC for quadruped robots. Gradient-based (via acados) or sampling-based in #JAX. Plus, multiple robots, multiple terrains, and multiple gaits! Link: github.com/iit-DLSLab/Q...

Hot new paper for bimanual manipulation of humanoid hands! bidex-teleop.github.io Whole setup costs <30k ; most affordable, for now :)

Teleop of in-home robot using a low-cost setup (all open sourced soon)

DinoV2 is without a doubt one of the most important Self Supervised Learning (SSL) methods right now. But training it takes 32 80Gb GPUs which is not easy to come by for small labs. What if we could train a comparable high-res model on 24Gb of VRAM? That's what I hope to show you here soon!🤞🧵 #mlsky

Since AI is hogging all the hype, I wanted to put together a starter pack specifically for roboticists focused on open-source and education. If you have recommendations, including yourself, let me know! go.bsky.app/5p6RBnm

This guy pretty much sums up how I feel about robots after a month of work travel.

If you're into #Robotics #Robots #Robot you might like this starter pack: go.bsky.app/5CNUmby The feed searches the above tags, so if you use them your skeet will appear. If you'd like to be added to the pack then comment here.

Model Predictive Control (MPC) for legged robots is limited by model complexity. In our recent IROS24 paper, we show how partitioning the robot into smaller subsystems, each controlled in parallel by a faster dedicated MPC, can achieve complex whole-body motions more efficiently. 1/2