⚠️ Important Notice: This website is now archived and will no longer be updated.

Welcome to the AE-353!

This website supports the course AE 353: Aerospace Control Systems taught by Prof. Melkior Ornik in Fall 2024 at the University of Illinois at Urbana-Champaign.

⚠️ Important Notice: Canvas will be the primary platform for all course communications. It is essential that students are enrolled in the correct Canvas page and have their notifications set up to receive announcements promptly. All crucial updates and announcements will be delivered through Canvas. This website is intended as a supplementary resource, primarily serving as a wiki for project-related information.

Course Description

AE 353 is a modern, challenging aerospace take on a standard undergraduate controls course. It seeks to use formal mathematical methods to answer a fundamental engineering question: How can I ensure that the system does what I want it to do?

On a more technical level, the primary objective of the course is to expose students to the notion of continuous-time, continuous-space controlled system dynamics, and explore the design of control signals to drive the system to a desired outcome. We will do so by formulating state space representations, relating the system’s control inputs, states, and outputs through a set of ordinary differential and algebraic equations. Even when such relationships are simple, performing control design in such a representation will require us to develop the method for computing a solution to a system of ordinary differential equations: the notion of a matrix exponential plays a significant role. Using a significant amount of linear algebra, we will then devote a large part of the course to discussing some of the central questions of control theory: can I drive the system to a particular state? (controllability), how can I do it in the quickest or cheapest way possible? (optimal control), and can I do it even if I don’t know everything about the system at every given time? (observability). This year, we will also make an effort to expose students to some relevant recent research in robotics, providing insight into the high-level purpose of the techniques that we are developing in class. Finally, we will connect the developed machinery of modern state-space-based control design to a classical method of frequency-domain-based control design – a “dual” of the state space representation that often simplifies control design methods, but pays the price of reduced applicability to more complicated dynamics models and control objectives.

Course Staff

Professor: Melkior Ornik (mornik@illinois.edu)

Teaching assistant: Gokul Puthumanaillam (gokulp2@illinois.edu)