NVIDIA Blog Robotics
Explores where AI, simulation, and GPU power collide in robotics and autonomous systems
For Robotaxis, Safety Must Be Built In, Not Bolted On
Robotaxi services are moving from prototypes to commercial fleets, with new collaborations announced at NVIDIA GTC Taipei. Uber and Autobrains, Foxconn, VinFast and HUMAIN will deploy vehicles on NVIDIA DRIVE Hyperion, expanding operations across Europe, Asia and the Middle East. NVIDIA introduced the Halos Operating System, a safety‑certified OS, SDK and application layer designed to meet ISO 26262 ASIL‑D standards. The accompanying Halos Safety Evaluation Framework provides large‑scale validation tools to prove reliability before public road deployment.
NVIDIA and LG Group Build an AI Factory to Advance Physical AI, Mobility and AI Infrastructure
NVIDIA and LG Group announced an AI factory that will combine NVIDIA’s full‑stack AI platform with LG’s expertise in consumer electronics, robotics, and data‑center technologies. The joint effort will deliver accelerated computing infrastructure for training, simulating, and deploying AI across...
NVIDIA and Doosan Group Collaborate to Advance Physical AI and AI Factory Infrastructure
NVIDIA and South Korea’s Doosan Group have deepened their partnership to push physical AI, robotics, and AI‑factory infrastructure across Doosan’s four business units. The collaboration blends NVIDIA’s full‑stack accelerated computing, including DSX, MGX, Isaac Sim, and Jetson Thor, with Doosan...
NVIDIA Research Unlocks Advanced Grasping, Smarter Autonomous Driving and Agent Training at Scale
At CVPR, NVIDIA Research unveiled three breakthrough papers that showcase how training at massive scale can produce foundation models capable of zero‑shot generalization. GraspGen‑X, built on 2 billion simulated grasps, can generate reliable grasp poses for any robotic gripper without per‑device...
NVIDIA Factory Operations Blueprint Gives Factories a New AI Brain
NVIDIA unveiled the Factory Operations Blueprint (FOX) at GTC Taipei, a reference design that lets manufacturers build autonomous factory manager agents to monitor, reason over, and orchestrate real‑time production data. The blueprint runs on the DGX Station powered by the...
How Cosmos 3 Helps Physical AI Think Before It Acts
NVIDIA’s Cosmos 3 platform merges live factory data with Omniverse‑powered simulations, letting physical AI evaluate actions before execution. The system underpins the Factory Operations Blueprint, advances robotics from simulation to real‑world deployment, and supports a simulation‑first design cycle. At Hannover Messe 2026, NVIDIA...
NVIDIA Research Advances Robotics From Simulation to the Real World
At ICRA, NVIDIA Research presented eight papers demonstrating how simulation‑to‑real transfer is becoming a core enabler for reliable, generalizable robotics. New frameworks such as ScheduleStream, COMPASS, Grasp‑MPC, SPARR and PEEK show dramatic gains in multi‑arm coordination, navigation across robot bodies,...
Into the Omniverse: Manufacturing’s Simulation-First Era Has Arrived
NVIDIA’s Omniverse platform, anchored by the OpenUSD‑based SimReady standard, is ushering in a simulation‑first era for manufacturing. By delivering physics‑accurate, photorealistic digital twins, companies can train and validate AI models before any physical hardware exists. Real‑world examples include ABB Robotics...
NVIDIA and Partners Showcase the Future of AI-Driven Manufacturing at Hannover Messe 2026
At Hannover Messe 2026, NVIDIA and partners demonstrated a full suite of AI‑driven manufacturing solutions, from accelerated computing infrastructure to real‑time digital twins and autonomous robots. The Industrial AI Cloud, built by Deutsche Telekom on NVIDIA hardware, provides a sovereign,...
National Robotics Week — Latest Physical AI Research, Breakthroughs and Resources
National Robotics Week highlights NVIDIA’s push to bring artificial intelligence into the physical world, showcasing breakthroughs in robot learning, simulation, and foundation models. The company’s platforms for simulation, synthetic data, and AI‑powered learning enable faster transition from virtual training to...
Into the Omniverse: Physical AI Open Models and Frameworks Advance Robots and Autonomous Systems
NVIDIA unveiled a comprehensive open‑source physical AI stack at CES 2026, linking high‑fidelity simulation, synthetic data generation, cloud orchestration and edge deployment through OpenUSD and Omniverse. The suite includes Isaac Lab‑Arena, Cosmos world models, Alpamayo vehicle models and OSMO orchestration,...
Japan Science and Technology Agency Develops NVIDIA-Powered Moonshot Robot for Elderly Care
Japan’s Science and Technology Agency is advancing its Moonshot program to field AI‑driven caregiving robots by 2050. The AIREC family, including the mobile Dry‑AIREC and data‑focused AIREC‑Basic, integrates multiple NVIDIA GPUs and Jetson Orin NX modules for edge AI processing....
Steel, Sensors and Silicon: How Caterpillar Is Bringing Edge AI to the Jobsite
The request does not contain the article text needed to generate a summary; instead it contains a meta‑request about cleaning up article HTML.
Into the Omniverse: OpenUSD and NVIDIA Halos Accelerate Safety for Robotaxis, Physical AI Systems
NVIDIA announced that the OpenUSD Core Specification 1.0 is now public, establishing a unified data model for high‑fidelity simulation pipelines. Powered by this standard, Omniverse libraries deliver SimReady assets that integrate seamlessly with Isaac Sim for robot and autonomous‑vehicle testing....
Cheers to AI: ADAM Robot Bartender Makes Drinks at Vegas Golden Knights Game
Richtech Robotics has deployed ADAM, an Automated Dual‑Arm Mixologist, at the T‑Mobile Arena during Vegas Golden Knights games. The robot, powered by NVIDIA’s Isaac simulation suite and Jetson AGX Orin edge AI, was trained in virtual environments before serving drinks...