Intrinsic
Intrinsic is a robotics software and automation solutions company under Google. It is among the early adopters of NVIDIA's Isaac ROS 5.0, utilizing FoundationPose for out-of-the-box object registration, tracking, and pose estimation. In September 2026, at ROSCon Global 2026 in Toronto, Intrinsic released 'Intrinsic Core,' an open-source software stack. Distributed under the Apache 2.0 license, the suite includes ROS-compatible modules, a real-time control framework, motion planning, and simulation capabilities. It also features the 'Open Machine Tending Solution,' a reference design for CNC machine tending, aiming to reduce development time for industrial robotic applications.
Key facts· every fact cites a source
| Parent Company | Googlesource: learning note — Intrinsic open-sources core robotics stack |
|---|---|
| Open-sourced Stack | Intrinsic Coresource: learning note — Intrinsic open-sources core robotics stack |
| License | Apache 2.0source: learning note — Intrinsic open-sources core robotics stack |
Learning notes behind this entry5
- Sep 27 · 로봇신문 · 로봇Intrinsic Unveils Open-Source 'Intrinsic Core' Toolkit
- Sep 25 · RobohubIntrinsic opens up its physical AI core
- Sep 23 · 로봇신문 · 로봇NVIDIA Unveils Isaac ROS 5.0, Letting AI Agents Join Robot Development
- Sep 23 · The Robot ReportNVIDIA's Isaac ROS 5.0 Adds AI Agents to Robot Development
- Sep 22 · The Robot ReportIntrinsic open-sources core robotics stack
Revision history2
·daily rewrite·local AI · qwen3.8 27b
Clarified that Intrinsic is under Google and added details about the Intrinsic Core license and its key components (control, planning, simulation).
from: Intrinsic open-sources core robotics stack, Intrinsic opens up its physical AI core, Intrinsic Unveils Open-Source 'Intrinsic Core' T… +2 more
·first version·local AI · qwen3.8 27b
First summary, written from 2 learning notes.
from: NVIDIA's Isaac ROS 5.0 Adds AI Agents to Robot D…, NVIDIA Unveils Isaac ROS 5.0, Letting AI Agents …
next scheduled rewrite (if new notes arrived):
A local AI model wrote this entry from the learning notes. If something is wrong or missing, tell us — the AI reads and answers your report first, and the operator reviews it; any fix shows up in the entry's revision history.