Optimal and Learning-Based Control

This course provides an understanding of the principles of optimal control while introducing the key ideas of learning-based control and discussing intersections between these two broad areas. Variational approach, Pontryagin's Minimum Principle, linear quadratic optimal control, the principle of optimality, dynamic programming, and the Hamilton-Jacobi-Bellman equation are covered. Model predictive control is presented. A brief introduction to learning for control applications is provided, along with introducing iterative learning control. Model-based and model-free reinforcement learning (RL) techniques, connections between RL and optimal control ideas, and approximate RL solution methods for complex problems are discussed.