Teaching

Learning Objectives

By the end of this course, students will be able to:

• identify and describe fundamental concepts in control (LTI system model, feedback control and stability, PID), optimization (LQR and MPC), and estimation (Kalman filter, recursive filters);

• implement basic algorithms for solving robot control problems;

• describe the core ideas behind more advanced nonlinear control concepts (inverse dynamics control, impedance control, and quadratic programming control) specific to articulated robots; and

• comprehend and summarize related research papers in the core area "Thinking about Actions".

Course Schedule

The course schedule is organized around four main topics: feedback control, state estimation, optimization-based control, and articulated robot control. 

Introduction

Lecture 1 | Course logistics, What control is about 

Feedback Control

Lectures 2-4 | LTI systems, feedback control and stability, PID

• What control is about

• Fundamental control model (LTI SISO systems)

• Feedback control and stability

• Application: PID controllers

State Estimation

Lectures 5-7 | Observability, Kalman filter, recursive estimation

• Kalman filter fundamentals

• Beyond Kalman filter: extended and unscented variants

• Application: Pedestrian dead reckoning with MEMS IMU

Optimization-Based Control

Lectures 8-10 | LQR and MPC

• Linear quadratic regulator (LQR)

• Model predictive control (MPC)

• Application: Autonomous lane keeping

Articulated Robot Control

Lectures 11-13 | Articulated robot control concepts

• Precision tracking with inverse dynamics control

• Human-robot collaboration with impedance control

• Application: Humanoid walking

• Course summary and bridge to learning-based approaches

Learning Resources

The course content draws from two textbooks:

• Feedback Systems: An Introduction for Scientists and Engineers

by Karl Johan Åström and Richard M. Murray (2010)

• Model Predictive Control: Theory, Computation, and Design

by James B. Rawlings, David Q. Mayne, and Moritz M. Diehl (2017)

Additional resources include:

• Control Systems Engineering 

by Norman S. Nise (2020)

A comprehensive, undergraduate-level introduction to design and analysis of feedback systems that emphasizes accessibility and practical application examples.

Programming Tools

Students use either MATLAB or Python for programming problems in the assignments. Students are responsible for ensuring they have access to and proficiency with their chosen environment. Resources include official documentation, CMU library resources, and online learning platforms focusing on scientific computing and numerical methods.

Assessment

Students' comprehension of control fundamentals and their practical application skills are assessed through three assignments. These assignments combine implementation challenges with reviews of relevant research papers. Example tasks include:

• implementing a Kalman filter to estimate the inertial state of a quadcopter, with recent research papers provided to guide the work,

• implementing a linear MPC controller for autonomous lane keeping, while using research papers to explore more advanced, nonlinear MPC for the same application.