NVIDIA released Isaac ROS 5.0 on September 22, bringing AI-agent workflows into the open-source ROS robotics ecosystem while expanding GPU acceleration across Jetson-powered robots.

The release arrived at ROSCon in Toronto and targets a practical problem in physical AI: developers need to build increasingly complex perception and manipulation software while also keeping those workloads fast enough to run on real robots.

AI agents can now help build robot software

Isaac ROS 5.0 adds agent-ready documentation and reusable skills that allow AI coding agents to assist with setup, migration and robotics workflows. NVIDIA says a new FoundationStereo fine-tuning skill can help adapt stereo-perception models to a developer’s cameras and environment.

The embedded NVIDIA Developer video is in English and demonstrates Isaac ROS acceleration on Jetson-class robotics hardware, so translated voice audio is not required.

FoundationPose runs up to 5.5 times faster

NVIDIA says the new agent-ready FoundationPose inference library can perceive and track the position and orientation of objects up to 5.5 times faster. Pick-and-place is also available as a standalone agent-ready skill combining detection, depth estimation and pose output.

The release adds ROS Lyrical and Ubuntu 24.04 support. NVIDIA also contributed a standard data-handling interface to ROS designed to improve movement of data across computing hardware, with CUDA providing a GPU-accelerated implementation.

Robot makers are already integrating Isaac ROS

Universal Robots is using Isaac ROS in its AI Accelerator SDK. ROBOTIS is integrating it into the AI Worker robot for vision-guided picking, placement and alignment. FieldAI is integrating Isaac ROS on Jetson devices for robot foundation models that can run locally without relying on cloud connectivity.

Noble Machines is also using Isaac ROS on Jetson to accelerate development of general-purpose industrial robots. Together, these deployments show how edge AI hardware and open robotics software are converging into a common physical-AI development stack.

Free and open source

Isaac ROS 5.0 is available now as free, open-source software. NVIDIA says the broader ROS ecosystem includes nearly 1.3 million users, giving the release a large developer base for GPU-accelerated robotics applications.

Sources: NVIDIA’s September 22, 2026 Isaac ROS 5.0 release materials and NVIDIA Developer technical documentation. Performance figures are NVIDIA-reported and may vary by robot, model and hardware configuration.

BitcoinVersus.Tech

Advertisement: Follow BitcoinVersus.Tech for independent coverage of Bitcoin, mining hardware, artificial intelligence, robotics, data centers, semiconductors, energy, gaming and emerging technology.

BitcoinVersus.Tech Editor’s Note & Disclaimer:
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. BitcoinVersus.tech provides news, technical analysis and educational information and does not provide financial or investment advice. Verify technical specifications and performance claims independently before making purchasing, deployment or investment decisions. If you would like to support to help further secure the integrity of our research initiatives, please donate here: 3C9o19EH5HSiwEPyCTmEKzxhNCbo2X6TTb

Related BitcoinVersus.tech coverage: Qualcomm PickNik robotics · Agility Digit 5 · PAL Robotics TIAGo Pro · QBit robot SoC