Overview: This is an advanced seminar course on AI-enabled robotics. The key question we will look at is how to build generalist robots that constantly learn, act and improve through natural interactions with the environment. We will study how robots perceive and model the complex world, make plans and decisions, and robustly adapt to various environmental conditions. Students will read, present and discuss the latest research on robot learning which involve areas in robot perception, manipulation, navigation, motion and task planning, robot and sensor design, and multi-robot systems. Throughout this course, students will also conduct a research-level project on robot learning topics.
This course is designed for graduate students and ambitious undergraduate students. If you are unclear whether you meet the prerequisites, please consult the instructor in advance. Undergraduate students must obtain explicit approval from the instructor.
Familiarity with calculus, statistics, and linear algebra.
Practical experience in computer programming. Proficiency in Python is required.
Coursework or equivalent experience in basic machine learning, computer vision and robotics.