NVIDIA has released several updates about its partners and its powerful AI ecosystem over at the ROSCon conference in Singapore, drawing together the global community behind the Robot Operating System (ROS), which is the world’s most widely adopted open framework for building robots, in an effort to accelerate and innovate the robotic industry through open-source collaborations.
To start, Team Green will be supporting the Open Source Robotics Alliance’s (OSRA) new Physical AI Special Interest Group, which focuses on real-time robot control, accelerated AI processing, and improved tools for autonomous behavior so that it can be adopted quickly and by devs around the world.
They are also contributing GPU-aware abstractions directly into ROS 2, enabling the framework to intelligently manage diverse processor types from CPUs to integrated and discrete GPUs for faster, more consistent performance. The contribution not only enhances efficiency but also future-proofs ROS against rapid hardware evolution across the robotics industry.
To further aid developers, NVIDIA introduced Greenwave Monitor, an open-source performance diagnostic tool that helps identify bottlenecks and streamline robot optimization. Complementing this, the company rolled out Isaac ROS 4.0, a new suite of GPU-accelerated libraries and AI models for the Jetson Thor platform, giving developers access to CUDA-optimized workflows for robot manipulation and mobility.
Partners in the robotics field also made strides with NVIDIA technology. For example, AgileX Robotics leverages NVIDIA Jetson modules for AI-driven autonomy and vision, while Canonical is offering a fully open observability stack for ROS 2 devices on Ubuntu, optimized for Jetson AGX Thor. Ekumen Labs has integrated NVIDIA Isaac Sim for high-fidelity testing and photorealistic simulation, and Intrinsic is combining Isaac foundation models and Omniverse simulation tools in its Flowstate platform to advance robot-grasping and digital twin visualization.
Other notable collaborations include KABAM Robotics, whose Matrix robot runs on Jetson Orin and Triton Inference Server for outdoor facility management; Robotec.ai, which is co-developing a new ROS simulation standard integrated into Isaac Sim; ROBOTIS, whose AI Worker robot uses the Isaac GR00T N1.5 model for enhanced autonomy; and Stereolabs, which offers full Jetson Thor compatibility through its ZED cameras and SDK for spatial AI vision.
From open frameworks and simulation tools to production-ready hardware, NVIDIA’s growing ecosystem continues to empower developers to build, test, and deploy the next generation of physical AI and autonomous robots.
