Building AI robots costs millions of dollars. Most engineers spend years just to build basic prototypes. Today, we launched NVIDIA Isaac Sim on Open GRID - making advanced robotics possible from your laptop. Here's how we're transforming a $200B industry:

Bringing humanoids, manipulators, and legged robots to Open GRID - powered by Isaac Sim! Just through your browser, choose from a variety of robots, equip them with state of the art AI such as VLMs, CV models; and build complex skills for any robot form factor in minutes.

Drone giant DJI didn't appear overnight. Their decade-long dominance stems from CEO Frank Wang. Wang's research and his eventual 2011 thesis defined the blueprint that revolutionized quadcopters forever. Let's revisit this thesis here: ⬇️

Data is the fuel for building any AI system. Let’s break down how you can generate data for Robot AI training seamlessly. Here is my GRID notebook that you can use to follow along: grid.scaledfoundations.ai/shared/14a128b…

Looking at the robotics talent landscape: Our analysis shows approximately one engineer capable of working on robotic systems for every 20 ML specialists. This poses interesting challenges for scaling physical AI applications.

Scaling Signal Temporal Logic (STL) for real-time robotics and AI! Blog: sca.fo/stlcg-plus-plus STLCG based methods are largely restricted to small-scale, offline applications due to it’s computational costs. Introducing STLCG++, a new approach inspired by causal masking used

Thrilled to share our work on accelerating neurosymbolic methods for robotics! STLCG++ encodes logical constraints in a computational graph, achieving 100-1000× faster quantitative evaluation—enabling real-time symbolic reasoning for traj opt, policy learning, & more. 🧵 [1/n]

Hello Robots, Roboticists, Engineers, We believe general-purpose robot intelligence will emerge from AI skills—composable capabilities that sense, reason, and act. These are not tasks. They are not demos. They are capabilities—learned, reusable, and ever-adapting across form

Scaled Foundations is now General Robotics. We’re building general-purpose intelligence for every robot. Across any scenario in the physical world. Blog: generalrobotics.company/post/general-p…

Big news today - we’re now General Robotics! This reinforces our mission towards truly general robot intelligence, and more importantly, our belief that general intelligence emerges from modular compositions of many skills. GRID is the home for AI skills and sims across robots.

Humans fundamentally combine skills to solve complex tasks. Robots should too. 💻 grid.generalrobotics.dev 🔗 generalrobotics.company/post/general-p…

AI skills for every Trossen robot, anywhere—through GRID. Excited for what this collaboration unlocks: making robots useful, fast—for all Trossen and GRID users. More this week 👀

Hello Roboticists, Trossen Robots are now live on Open GRID. With just a few lines of code, any Trossen arm — old or new — can tap into AI skills from the cloud. One example: Teleoperation via API → data collection on Isaac Sim → playback on real hardware 👇🏻 Making Trossen

🤖 GRID Feature Release: Heterogeneous multi-agent control is now live on GRID Enterprise. As requested, all AirGen users can now coordinate mixed robot fleets — and adapt AI skills across form factors with ease. #robotics #AI #multiagent

One skill. Any arm. Pick and Place. In video: UR5e, WidowX #robotics #AISkills #manipulation

Generalization is the hallmark of intelligence - robots should adapt, not memorize. One skill, multiple robots, multiple objects.

Real → Sim → Robot — fast. With Gaussian splats, GRID brings physical assets into simulation tools like AirGen and NVIDIA Isaac Sim, enabling AI skills to be tested in sim and deployed to real robots in minutes. #Robotics #PhysicalAI #Simulation

95% of GenAI pilots fail — MIT Media Lab Robot AI is even more challenging. The 5% that succeed? They deploy. Experience AI deployed on your robots today! In video: Pick and place for Universal Robots forbes.com/sites/andreahi…

The power of generative models — now embodied in humanoids. Announcing DreamControl –– After a year-long research effort at General Robotics — we present a scalable framework for whole-body humanoid control that fuses diffusion priors with reinforcement learning to unlock

General Robotics is bringing natural whole-body skills on real robots!!!