Building Robotics Applications with Ryzen AI and ROS 2

2026-02-09 · Source: AMD ROCm Blogs · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Software Development & Engineering · Depth: Intermediate, medium

Summary

AMD showcases deploying power-efficient Ryzen AI perception models with ROS 2 on the Ryzen AI Max+ 395 (Strix-Halo) platform, which features an NPU and iGPU. The demonstration utilizes the Ryzen AI CVML Library for efficient model deployment and integrates with ROS 2's publisher/subscriber model. The provided code, available on GitHub in the AMD Ryzers repository, was initially presented at ROSCon'25. The setup involves installing the Ryzers framework and configuring a Docker container with XDNA drivers, Ryzen AI CVML, and ROS 2. A custom ROS 2 node wraps CVML features like depth estimation, face detection, and face mesh, with outputs visualized using standard ROS 2 nodes and `web_video_server`.

Key takeaway

For AI Engineers developing robotics applications on AMD Ryzen AI platforms, this guide demonstrates a clear path to integrating power-efficient perception models. You should leverage the Ryzers framework to streamline ROS 2 and CVML setup, enabling rapid prototyping and access to the extensive ROS ecosystem. Consider offloading perception tasks to the NPU to optimize performance and power consumption in your edge robotics deployments.

Key insights

Integrate Ryzen AI perception models into ROS 2 robotics applications for power-efficient edge processing.

Principles

• ROS 2 standardizes robotics application development.
• NPUs offload perception tasks for efficiency.

Method

Set up a Dockerized Ryzers environment with ROS 2 and Ryzen AI CVML, build a custom CVML ROS 2 node, publish video frames, launch the perception node, and visualize outputs using standard ROS 2 tools.

In practice

• Use `ryzers` for simplified ROS 2 and CVML setup.
• Wrap CVML models in ROS 2 nodes for ecosystem access.
• Verify NPU utilization with `xrt-smi examine`.

Topics

• Ryzen AI
• ROS 2
• NPU Acceleration
• Perception Models
• Robotics Applications

Code references

• AMDResearch/Ryzers
• amdresearch/ryzers

Best for: Robotics Engineer, AI Engineer, Software Engineer

Related on AIssential

• Your AI Setup Needs This — AMD Ryzen AI Processors integrate NPU, GPU, and CPU for enhanced on-device AI performance in business laptops.
• From Factory Precision to Intelligent Robots: How Real-World Performance Shapes Industrial Edge AI — Production-ready industrial edge AI demands deterministic performance and integrated, concurrent compute for diverse workloads.
• From AI Agents to Faster Kernels: Ben Burtenshaw & Felix LeClair (AI Plumbers #2) — AI agents can generate optimized kernels, but structured "skills" and robust infrastructure are crucial for widespread adoption and benefit.
• How Jaiveer Singh Is Helping Robots — and Developers — Move Faster — NVIDIA Isaac ROS provides modular, open-source, GPU-accelerated tools for rapid robotics development, fostering confidence and collaboration.
• CES 2026: Everything AMD revealed at the show — AMD's CES 2026 strategy focuses on an "AI everywhere" approach, spanning data center to edge with new silicon and platforms.
• AI giveth and AI taketh CPU — AMD's AI strategy leverages heterogeneous computing and modular chiplet designs for performance, efficiency, and adaptability across diverse workloads.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AMD ROCm Blogs.