About the course
This course is a practical, deep-dive journey into the world of End-to-End Learning for robotics. Instead of manually programming every joint movement with rigid code, you will learn how to "teach" the Niryo Ned2 cobot using LeRobot, the state-of-the-art framework from Hugging Face.
By the end of this program, you will have moved beyond traditional automation into the realm of Physical AI, where robots learn complex behaviors through demonstration and imitation.
Who Is This For?
This course is designed for those who want to stop theorizing about AI and start seeing it move in the physical world:
-
Robotics Engineers & Students: Looking to integrate modern ML workflows into their existing hardware setups.
-
AI Practitioners: Eager to move beyond computer vision or NLP and explore "Embodied AI" on real-world devices.
-
Makers & Educators: Who want a structured path to using professional-grade open-source tools like LeRobot.
-
Prerequisites: A basic comfort level with Python and a Niryo Ned2 robot (or access to one).
Course Content:
Part 1: AI applied to robotics & LeRobot Framework
Part 2: Interfacing Niryo Ned2 and LeRobot
Part 3: Create your first DatasetPart 3.2: Record your first dataset
Part 4: Train your own AI modelPart 4.2: Training your first own policy
What you'll learn
- The LeRobot Ecosystem: Understand how to navigate the framework that is standardizing AI for robotics.
- Hardware Interfacing: Learn the specifics of linking the Ned2 with LeRobot.
- Data Collection & Imitation Learning: Master the art of recording high-quality datasets through teleoperation to teach your robot new tasks.
- Policy Training: Move from raw data to a trained neural network (policy) that dictates the robot's movements.
- Asynchronous Deployment: Real-world testing using async inference to ensure smooth, responsive robot behavior during task execution.
- Technical Troubleshooting: Gain the "under-the-hood" knowledge to debug common hardware-software communication issues.
Course content
About the Author
Enrolment options
Introduction to Imitation Learning AI interfacing Niryo and LeRobot
- Content Type: Project
- Programming: Python
- Equipment: Ned2
This course is a practical, deep-dive journey into the world of End-to-End Learning for robotics. Instead of manually programming every joint movement with rigid code, you will learn how to "teach" the Niryo Ned2 cobot using LeRobot, the state-of-the-art framework from Hugging Face.
By the end of this program, you will have moved beyond traditional automation into the realm of Physical AI, where robots learn complex behaviors through demonstration and imitation.
Who Is This For?
This course is designed for those who want to stop theorizing about AI and start seeing it move in the physical world:
-
Robotics Engineers & Students: Looking to integrate modern ML workflows into their existing hardware setups.
-
AI Practitioners: Eager to move beyond computer vision or NLP and explore "Embodied AI" on real-world devices.
-
Makers & Educators: Who want a structured path to using professional-grade open-source tools like LeRobot.
-
Prerequisites: A basic comfort level with Python and a Niryo Ned2 robot (or access to one).
Course Content:
Part 1: AI applied to robotics & LeRobot Framework
Part 2: Interfacing Niryo Ned2 and LeRobot
Part 3: Create your first DatasetPart 3.2: Record your first dataset
Part 4: Train your own AI modelPart 4.2: Training your first own policy
- Enrolled students: 10
- Content Type: Project
- Programming: Python
- Equipment: Ned2
This course includes
Resources