University of Tokyo · Japan
Physical AI: Robot Control and AI Development
This is a practical, 7-session online course focused on "Physical AI," the intersection of robotics and artificial intelligence. Designed for students and professionals with a foundational understanding of machine learning and deep learning, the program aims to equip participants with the skills to "move robots using data." Through hands-on exercises using LeRobot in a simulated environment, you will learn to systematically build, train, and evaluate robot foundation models and control systems. The curriculum covers data collection, reinforcement learning for locomotion, utilization of learning data for scaling, and challenges in real-world deployment, offering a comprehensive understanding from foundational concepts to practical applications.
The course is developed and supervised by the Matsuo-Iwasawa Laboratory at the University of Tokyo, which has over 10 years of experience in nurturing data scientists and deep learning professionals, having trained over 70,000 individuals in 2025.
While the AI algorithm details are not covered in-depth, it provides a strong foundation for those interested in advanced topics, with references to other specialized courses.