Sudharshan Suresh

I'm a featured interview in our latest behind-the-scenes release! We break down the ML and perception that drives the whole-body manipulation behaviors from last year. It starts with a neat demo of Atlas's range-of-motion and our vision foundation models. youtu.be/oe1dke3Cf7I?si…

Honored to be a part of the RSS Pioneers cohort this year - look forward to catching up with folks in Los Angeles! 🤖

So excited for this!!! The key technical breakthrough here is that we can control joints and fingertips of the robot **without joint encoders**. Learning from self-supervised data collection is all you need for training the humanoid hand control you see below.

TRI's latest Large Behavior Model (LBM) paper landed on arxiv last night! Check out our project website: toyotaresearchinstitute.github.io/lbm1/ One of our main goals for this paper was to put out a very careful and thorough study on the topic to help people understand the state of the…

Current robot policies often face a tradeoff: they're either precise (but brittle) or generalizable (but imprecise). We present ViTaL, a framework that lets robots generalize precise, contact-rich manipulation skills across unseen environments with millimeter-level precision. 🧵

I'll be speaking at the RSS Dexterous Manipulation Workshop tomorrow, discussing our recent work with Atlas!

Today we're excited to share a glimpse of what we're building at Generalist. As a first step towards our mission of making general-purpose robots a reality, we're pushing the frontiers of what end-to-end AI models can achieve in the real world. Here's a preview of our early…

How to learn dexterous manipulation for any robot hand from a single human demonstration? Check out DexMachina, our new RL algorithm that learns long-horizon, bimanual dexterous policies for a variety of dexterous hands, articulated objects, and complex motions.

Check out the awesome follow-up to sparsh by @akashshrm02 and others - self-supervised learning for a variety of downstream tasks with tactile skins!

New work from the Robotics team at @AIatMeta . Want to be able to tell your robot bring you the keys from the table in the living room? Try out Locate 3D! interactive demo: locate3d.atmeta.com/demo model & code & dataset: github.com/facebookresear…

📢Sonata: Self-Supervised Learning of Reliable Point Representations📢 Meet Sonata, our"3D-DINO" pre-trained with Point Transformer V3, accepted at #CVPR2025! 🌍: xywu.me/sonata 📦: github.com/facebookresear… 🚀: github.com/Pointcept/Poin… 🔹Semantic-aware and spatial…

Atlas is demonstrating reinforcement learning policies developed using a motion capture suit. This demonstration was developed in partnership with Boston Dynamics and @rai_inst.

Check out our latest work, VGGT! ⚡️ It’s a fast transformer that predicts cameras, point maps, depth and 3D point tracks. VGGT + CoTracker significantly outperforms CoTracker in 2D point tracking.

Finally able to share what we've been working on the past few months 🤖🧠🖐️ We gave Gemini the ability to perceive, reason, and act in the physical world. deepmind.google/technologies/g… I want to especially highlight Gemini's embodied reasoning capabilities below

We’re designing Atlas to do anything and everything, but we get there one step at a time. See why we started with part sequencing, how we are solving hard problems, and how we’re delivering a humanoid robot with real value.

We’ve had fun testing the limits of MASt3R-SLAM on in-the-wild videos. Here’s the drone video of a Minnesota bowling alley that we’ve always wanted to reconstruct! Different scene scales, dynamic objects, specular surfaces, and fast motion.

🎅🤸

Introducing MASt3R-SLAM, the first real-time monocular dense SLAM with MASt3R as a foundation. Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map. With @eric_dexheimer*, @AjdDavison (*Equal Contribution)

CMU covered our work on the Robotics Institute website!