Topics Price war Tariffs Solid state ADAS Software Sales data Supply chain Battery recycling

Home /Technology

Technology 2 min read

NVIDIA launches Isaac ROS 5.0 to bring AI agents to robot development

NVIDIA unveiled Isaac ROS 5.0 at ROSCon, adding AI agent workflows, faster perception models, and broader Jetson support for robot developers.

NVIDIA has released Isaac ROS 5.0, a GPU-accelerated software suite built on the Robot Operating System (ROS), at the ROSCon conference in Toronto, Canada. The update aims to bring AI agent capabilities into the ROS developer ecosystem and expand the open-source physical AI library, according to Chinese automotive outlet Gasgoo.

Isaac ROS 5.0 adds support for ROS Lyrical and Ubuntu 24.04, giving developers a path to adopt the latest ROS platform. NVIDIA worked with the Open Source Robotics Alliance to provide a standardized data-processing interface for ROS Lyrical. The interface is open to the entire ROS community and is designed to help robot software run efficiently across different computing hardware, including GPUs, with CUDA cited as an example of GPU-accelerated implementation.

AI agent workflows and faster perception

In terms of AI agent workflows, the new version includes NVIDIA Isaac setup and operation skills, offering developers and agents reusable workflows. Agent-ready documentation makes it easier for AI agents to understand the relevant tools. On the technical side, a new FoundationStereo fine-tuning skill helps developers adapt stereo perception models to specific cameras and robot applications. FoundationPose, a foundation model for object pose estimation and tracking, now offers an agent-ready inference library that speeds up robot perception and tracking of object position and orientation by up to 5.5 times. Additionally, a pick-and-place workflow is now available as a standalone agent-ready skill.

Scalable deployment on Jetson

For hardware deployment, Isaac ROS 5.0 supports scalable compute from the entry-level NVIDIA Jetson Orin Nano to the high-performance Jetson Thor. The Jetson platform can run the physical AI technology stack at real-time performance on the edge, integrating ROS, accelerated perception and navigation, AI models, and application logic onto robots, providing developers with a clear path from development to deployment.

ROS is an open framework under Open Robotics, providing an open-source foundation for modern robot development. NVIDIA Isaac ROS brings accelerated computing, physical AI models, and production-ready libraries to nearly 1.3 million ROS users, helping developers build high-performance robot applications with open-source tools.