Advanced Robot Learning and Decision-Making

Master Wahlmodul, Advanced Control and Learning Course with Practical Implementations

This course covers advanced topics in robot learning and control, combining theoretical foundations with practical implementation to equip students with the skills to tackle real-world robotics challenges. The course emphasizes a systems-oriented approach, introducing modelling techniques for robot systems, optimal control methods, learning-based approaches, and reinforcement learning tailored to robotic applications. Students will learn how to model and control robotic platforms, integrate learned models into control frameworks, and address real-world challenges such as uncertainty, disturbances, and model inaccuracies.

Learning Objectives

At the end of the course, students are able to:

• Derive optimal controller equations for mobile robots and design learning-based approaches to cope with non-idealities such as model errors;
• Understand the strengths and limitations of various approaches, including optimal control, learning-based control, and reinforcement learning;
• Implement state-of-the-art algorithms, rapidly prototype new ideas, and apply them to their own problem domains.

Prerequisites

Motivation and interest in robotics, machine learning, and control, as well as good knowledge of calculus, probability theory, linear algebra, fundamentals of control, and Python programming (the programming exercises will be in Python). Additional knowledge and courses on the following topics will be beneficial: Dynamic systems, optimal control, classical reinforcement learning, basics of optimization, mobile robotics, and machine learning, deep learning (e.g., I2DL course). Moreover, our course, “Optimal Control and Decision Making,” which runs parallel to this one, serves as a good foundation but it is not mandatory.

Teaching and Learning Approach 

The core components of this course are

• Lectures (introducing the core concepts)
• In-Class Tutorials (working through examples and setups)
• Practical Sessions (hands-on implementation of algorithms, including both in-lecture and take-home components)

The following types of media are used

• Presentations (blackboard and slides)
• Handouts (lecture notes and assignments)
• Online Videos (additional tutorials)
• Python codebase and programming assignments

Evaluation

Oral exam and submission of programming assignments throughout the semester.