Introduction to Feedback Control Systems

12 units · Undergraduate

Spring: 12 units This undergraduate robotics course covers the fundamental concepts of feedback control and the central role feedback control plays in AI for robotics and robot learning. The course presents a unified approach to feedback control/policy design, optimal control, planning, and reinforcement learning. Topics include making models and robot dynamics, PID control, LQR control, state estimation including Kalman filters, insight from frequency domain techniques, path planning algorithms such as A* and Rapidly-Exploring Random Trees (RRT), dynamic programming, model-based reinforcement learning, and neural network approaches to reinforcement learning. Students will gain practical experience in designing and implementing feedback control both in simulation and on a real system. The course includes lectures, assignments, work with a real robot, and a project. By the end of the course, students will have a strong foundation in feedback control, and have gained insight into how to enable robots to make decisions and choose actions. Prerequisites: 15-122 and 21-122 Course Website: http://www.cs.cmu.edu/~cga/controls-intro/