ECE 789 - Model Predictive Control

This course tackles the problem of optimal control of dynamical systems with constraints. This is done through an optimization-based method called Model Predictive Control, or MPC. The course covers: 1) basic concepts of system theory, including state-estimation and hybrid systems, 2) convex optimization, constrained and unconstrained optimal control, 3) concepts of stability, reachability, invariant sets, 4) Model Predictive Control formulations (such as robust and hybrid MPC) and associated mathematical guarantees on robustness, optimality and recursive feasibility, 5) numerical methods for MPC. While the primary focus is on the underlying theory, the course will also cover modern software tools for implementing MPC.