Intrinsic open sources key parts of its platform for easier development
AI-summarised brief · reviewed before publication
Intrinsic, the AI‑robotics group acquired by Google, announced Intrinsic Core at ROSCon 2026 in Toronto. The open‑source package bundles a ROS‑compatible control framework, pose‑estimation, motion‑planning, grasp‑planning, simulation services and camera‑calibration tools, all released under an Apache 2.0 licence on GitHub. Intrinsic says the components are derived from software used in its own manufacturing deployments and are hardware‑agnostic, allowing developers to swap arms, sensors or grippers without rewriting drivers. The suite also includes NVIDIA FoundationPose for six‑degree‑of‑freedom object tracking and Gazebo‑based simulation for rapid testing. By providing pre‑configured building blocks, Intrinsic aims to cut the hundreds of hours developers currently spend coding robotic capabilities from scratch. The release targets large manufacturers and academic labs, promising faster prototyping and access to robotics technology.
💡 Why It Matters
- · By open‑sourcing its production‑grade stack, Intrinsic lowers the barrier for teams to deploy adaptive robots, accelerating real‑world automation beyond niche labs.
