Stanford Robotics Independent Study
A hands-on introduction to building AI-enabled robots.
2023 Teaching team:
Head teacher: Gabrael Levine (CS 2023, Stanford Student Robotics)
Head teacher: Jaden Clark (CS 2024, Stanford Student Robotics)
Lead TA: Mark Leone (ME 2024, Stanford Student Robotics)
Stuart Bowers (Hands-On Robotics)
Jie Tan (Google DeepMind)
Tingnan Zhang (Google DeepMind)
Adrien Gaidon (Toyota Research Institute)
Welcome to the course page for Stanford Student Robotics’ course in legged robots. The course has been taught at Stanford, UW, WashU, Foothills College, Brandeis, and George Mason, with plans to expand worldwide!
In the first six weeks, students will learn key robotics concepts like including motor control, forward and inverse kinematics, and system identification; as well as important embodied-AI concepts including reinforcement learning and simulation. Through weekly labs, students will build a pair of teleoperated robot arms with haptic feedback, program a robot arm to learn to move by itself, and most importantly, build and program an agile robot quadruped called Pupper (pictured above). In the last four weeks of the course, students will pursue an open-ended project using Pupper as a platform, such as teaching Pupper to walk using reinforcement learning, building a vision system so Pupper can play fetch, or redesigning the hardware to make the robot more agile.
Researchers from Google DeepMind and Toyota Research Institute will give guest lectures during the quarter on their work teaching robots new skills using reinforcement learning.
“Empowering robots with AI is essential to make them smart and useful in people’s daily life. It is one of the most important research directions in both academia and industry. This class teaches the most relevant skills, gives students hand-on experiences, and prepares them for a career in the area of AI and robotics.” - Jie Tan, Staff Research Scientist at Google DeepMind
Expected time commitment: 6 - 8 hours per week.
Estimated class size: 8 - 15 students
Prerequisites: CS106B or similar coding experience strongly recommended. Coding will be majority Python but some C++ (Arduino). Familiarity with the command line. Math 51 or CME 100 or equivalent understanding of gradients. No robotics experience necessary!!
Grading: Pass/Fail for 2 or more units. Grading based on participation.
Spring Quarter Faculty sponsor: Professor Karen Liu
- Lab 1 - Hello PID
- Lab 2 - Bad Robot Surgeon
- Lab 3 - Forward Kinematics
- Lab 4 - Inverse Kinematics
- Lab 5 - Reinforcement Learning
- Lab 6 - Pupper Assembly
- Lab 7 - Pupper Control and Simulation
- Final Project