Fall | Graduate | 12 Units | Prereq: 2.151 or permission of instructor
Introduction to robotics and learning in machines. Kinematics and dynamics of rigid body systems. Adaptive control, system identification, sparse representations. Force control, adaptive visual servoing. Task planning, teleoperation, imitation learning. Navigation. Underactuated systems, approximate optimization and control. Dynamics of learning and optimization in networks. Elements of biological planning and control. Motor primitives, entrainment, active sensing, binding models. Term projects. In person not required.
Fall 2020 Update: Fully Remote Classes - This subject is not taught every year. It is very much a graduate course in the sense that a lot of initiative (readings etc.) is expected from the students, the class is interactive, and a good mathematical background at the graduate level is assumed. It is typically a small class and some topics in the course description may be covered in more depth according to student interests, with particular emphasis this year to continuous-time and geometric understanding of machine learning algorithms in robotic contexts. Remote lectures will be mixed with video lectures and other video material. Grading will be mostly through term projects.