Robots For Humanity 🦾

At Robots For Humanity, we’ve redefined the way humanoid robots are trained. Using NVIDIA’s GR00T N1.5, we can now train a robot to perform a task just by ing a video of a human doing it. Here’s how we do it: 🎥 We start with a simple video of a repetitive task. 🧠 Using NVIDIA Robotics GR00T-Dreams, we generate synthetic data to enrich the training set. 🏭 We create a digital twin of the real environment in Isaac Sim. 🦾 Our robots train virtually, learning every movement and detail in a simulated world. 🔁 Once deployed physically, they operate with precision from day one — and keep improving through real-world . This process allows us to scale robot deployment, reduce integration time, and train with incredible speed and safety. We’re not just deploying robots. We’re building a system where robots learn like humans — by observing, simulating, and iterating. #RobotsForHumanity #GR00T #NVIDIA #IsaacSim #DigitalTwins #HumanoidRobots #SyntheticData #PhysicalAI #RobotTraining #IndustrialAutomation #AI

Robots For Humanity 🦾 featured in LA NACION 🇦🇷🤖 We’re honored to be featured by LA NACION, one of Argentina’s leading newspapers, for our mission to train and deploy humanoid robots from Córdoba to the world. At Robots For Humanity 🦾, we’re transforming how industries handle physically demanding and high-risk tasks. By combining humanoid robots with hyper-realistic digital twins—powered by platforms like NVIDIA Omniverse—we’re enabling seamless, efficient, and human-centered automation. Huge thanks to María Gabriela Ensinck for helping us tell our story and highlighting how technology from Latin America is shaping the future of robotics and industry 🌍🦾 #Robotics #Humanoids #DigitalTwins #Industry40 #Innovation #Startups #Córdoba #Argentina #LaNacion https://lnkd.in/dEp9MtR8

At Robots For Humanity 🦾 , we’re now training our humanoid robots using GROOT N1, NVIDIA Robotics latest foundation model for robotics. This breakthrough enables our robots to learn faster and more accurately inside hyper-realistic digital twins built in NVIDIA Omniverse, allowing for safer and more efficient deployment in real-world environments. 🎥 Over the coming weeks, we’ll begin publishing thousands of training videos of humanoid robots. These datasets will power our operating system, RobotsOS H1, which learns in real time and continuously enhances robot performance in the field. The robotic revolution has begun—and we’re training it. #RobotsForHumanity #NVIDIA #GROOT #Omniverse #RobotsOS #HumanoidRobots #DigitalTwins #FutureOfWork

Update: Figure at BMW Manufacturing Co., LLC At Robots For Humanity 🦾, we want to congratulate the team at Figure AI for the incredible milestone achieved with the deployment of Figure 02 at BMW’s Spartanburg plant, powered by the breakthrough intelligence of Helix. Seeing a humanoid robot perform real tasks on a production line is not just a technical achievement — it’s a powerful step toward a future where humans and robots collaborate safely, efficiently, and naturally. With Helix, Figure AI enables: - Continuous high-frequency upper body control. - Natural language understanding for complex actions. - Precise manipulation of unknown objects. - Multi-robot collaboration without t training. At Robots For Humanity, we celebrate every step that brings humanoid robotics closer to transforming industries, improving lives, and opening new opportunities for people. Congratulations to the entire Figure AI team — it’s inspiring to see the future being built! #Robots #Humanoids #Helix #FigureAI #BMW #Innovation #Automation #RobotsForHumanity

Mercedez Benz is now working with humanoid robots on its factory floor: Apptronik Apollo steps in. Standing 5’8” tall and weighing 160 lbs, Apollo is a general-purpose humanoid robot built for industrial environments. It can lift up to 55 lbs, deliver assembly kits, inspect vehicle components, and transport parts throughout different stages of the manufacturing process. Mercedes-Benz is adopting Apollo to address a growing challenge: the lack of reliable labor for physically demanding, repetitive, and low-skill tasks. 💬 As Jeff Cardenas, CEO of Apptronik, put it: “This is a model use case that we’ll see other organizations replicate in the months and years to come.” At Robots For Humanity 🦾, we celebrate every milestone that shows humanoid robots are no longer future tech—they are present-day solutions. And we’re building a world where these solutions become the standard in the most demanding industries. 🌍 Can you imagine your factory with humanoid robots working side by side with your team? #RobotsForHumanity #Apollo #MercedesBenz #Apptronik #HumanoidRobots #Industry40 #FutureOfWork #Automation #AI #Manufacturing #Robotics

Fourier Intelligence’s GR-1: Advancing Humanoid Robotics with NVIDIA Robotics GR00T N1 The GR-1 humanoid robot by Fourier Intelligence is now enhanced with NVIDIA’s GR00T N1 foundation model, enabling advanced multimodal reasoning and real-time action planning. This integration allows GR-1 to understand complex instructions and perform intricate tasks with human-like precision. GR00T N1’s dual-system architecture—combining a vision-language model with a diffusion transformer—empowers GR-1 to interpret its environment and execute fluid movements seamlessly. Trained on a vast dataset of real and synthetic data, GR00T N1 facilitates GR-1’s adaptability across diverse applications. At Robots For Humanity 🦾, we celebrate these breakthroughs as part of the ecosystem that is redefining the future of work. We’re excited to collaborate with technologies like Fourier GR-1 and GR00T N1 to deploy fully trained robots ready to operate in real industrial environments. This collaboration marks a significant step toward versatile, general-purpose humanoid robots capable of operating in dynamic real-world environments. #HumanoidRobotics #AI #GR1 #GR00T #NVIDIA #FourierIntelligence #RoboticsInnovation #Omniverse #RobotsForHumanity

🦾 | Figure on the move This is what the future looks like rolling down the highway: a full load of Figure 02 humanoids headed for their first shift. Each unit was trained in a hyper‑realistic digital twin, so they hit the factory floor with zero downtime and millimetre‑level precision. We’re no longer talking pilots—this is industrial scale. Ready to watch work evolve? Welcome to the next frontier. #Robotics #Automation #DigitalTwins #FutureOfWork #humanoidRobots

Introducing NVIDIA Robotics GR00T N1 — the world’s first open foundation model for humanoid robots With GR00T N1, NVIDIA sets a new benchmark for embodied AI. This cross-embodiment foundation model enables robots to reason and act using multimodal input — combining vision, language, and spatial context. Trained on a massive dataset including real-world capture, synthetic trajectories via GR00T Blueprint, and internet-scale video, GR00T N1 allows humanoids to: 🔹 Grasp and move objects with one or both arms 🔹 Transfer items hand-to-hand 🔹 Execute multi-step tasks with long-horizon reasoning 🔹 Generalize skills across manipulation, logistics, inspection, and more 🛠️ GR00T Workflows unlock a modular training pipeline: GR00T-Teleop: collect expert demonstrations via teleoperation GR00T-Mimic: scale from a few demos to thousands of synthetic motions GR00T-Gen: train in realistic simulation environments GR00T-Dexterity: pixel-to-action grasping with human-like dexterity GR00T-Mobility: adaptive locomotion using RL + imitation GR00T-Control: whole-body precision control libraries GR00T-Perception: VLMs + memory for enhanced situational awareness At Robots For Humanity 🦾, we’re integrating this framework into real-world industrial deployments. We create hyper-realistic digital twins using NVIDIA Omniverse, train robots with GR00T-based pipelines, and deploy humanoids that arrive ready to work — optimized, safe, and aligned with real factory tasks. The simulation-to-reality gap is closing. This is how we teach robots to learn. #GR00T #NVIDIAIsaac #EmbodiedAI #HumanoidRobots #RoboticsFoundationModel #DigitalTwins #ReinforcementLearning #SimulationToReality #RobotsForHumanity #Industry40

Congratulations to Artem Sokolov and the Humanoid team on the launch of the HMND 01 — redefining elegance for humanoid robots. With 4 hours of autonomy and a 1.5 m/s walking speed, it’s a powerful step toward seamless human-robot collaboration in real-world environments. From Robots For Humanity 🦾, we celebrate every advance that brings us closer to a future where robots amplify, not replace, human potential. #Humanoids #Robotics #FutureOfWork #AI #DigitalTwins #EleganceInEngineering

Walker S1 in Action: Redefining Work from Latin America At Robots For Humanity 🦾, we are already deploying the UBTECH Robotics Walker S1 humanoid robot by UBTECH Robotics in real industrial environments—bringing humanoid automation out of the lab and into the factory floor. In these photos, you can see the Walker S1: • Collaborating with an autonomous mobile robot (AMR) for parts inspection and logistics • Lifting crates inside a distribution center with precision and autonomy • Organizing bins at a workstation, adapting to existing human workflows These are real deployments, not simulations. We train Walker S1 in hyper-realistic digital twins using NVIDIA Omniverse, then deploy it fully prepared—reducing risk and accelerating value. From Argentina 🇦🇷 to the world, we’re proving that humanoid robotics is no longer science fiction—it’s an industrial reality. #HumanoidRobots #WalkerS1 #DigitalTwins #NVIDIAOmniverse #RobotsForHumanity #FutureOfWork #Industry40 #Automation #MadeInLatam