# Embodied AI & Robotics

> AI that acts in the physical world: robot foundation models like Physical Intelligence π0 and NVIDIA Cosmos, simulation and robot learning environments like NVIDIA Isaac Sim and MuJoCo, and middleware like ROS 2 connecting learning systems to real hardware.

8 tools in 3 subcategories. Web page: https://ailandscape.org/category/embodied-ai-robotics · Each tool also has a markdown version at /tool/{slug}.md

## Robot Foundation Models

- [Hugging Face LeRobot](https://ailandscape.org/tool/hugging-face-lerobot): Open-source robot learning models, datasets, and real-robot tooling
- [NVIDIA Cosmos](https://ailandscape.org/tool/nvidia-cosmos): World foundation models for physical AI development
- [Physical Intelligence π0](https://ailandscape.org/tool/physical-intelligence-0): Vision-language-action robot foundation model with open weights
- [Xiaomi Robotics XR-1](https://ailandscape.org/tool/xiaomi-robotics-xr-1): Xiaomi's open-source vision-language-action foundation model for robots

## Simulation & Training

- [MuJoCo](https://ailandscape.org/tool/mujoco): Physics engine for robotics, biomechanics, and RL research
- [NVIDIA Isaac Lab](https://ailandscape.org/tool/nvidia-isaac-lab): Open-source framework for robot learning built on Isaac Sim
- [NVIDIA Isaac Sim](https://ailandscape.org/tool/nvidia-isaac-sim): NVIDIA's Omniverse robot simulator for testing, training, and synthetic data

## Robot Middleware

- [ROS 2](https://ailandscape.org/tool/ros-2): Open-source robot operating system and middleware ecosystem

---

Part of [AI Landscape](https://ailandscape.org), an open map of the AI ecosystem. Index for AI assistants: https://ailandscape.org/llms.txt