Robotics

The course was previously taught by Herman Bruyninckx. I will teach the course in academic year 2023–2024.

Objectives

This course is an introduction to Intelligent Robotic Systems, i.e., machines that move (themselves and/or objects in their environment) and sense what is going on in their (immediate) neighbourhood, in order to achieve a given goal under uncertain environment conditions.

Applying AI techniques to a physical system poses challenges that are not apparent in other contexts. This course aims to teach how one casts an embodied-agent problem to a form that lends itself to an AI solution, for instance by choosing AI techniques, sensor/data representations and motor command schemes that are synergetic with one another.

This course will cover both “classical” AI techniques that are easily parametrized by an expert, and techniques that are learned from data. We will study the applicability of both approaches and discuss how to judiciously choose one or the other based on the nature of the task.

Since robotics is about integrating the best things from several research areas (mechanics, computer science, geometry, artificial intelligence, …), relationships with other courses often occur, but we avoid overlaps as much as possible. The students are intensively stimulated to think and discuss as a researcher.

During the course, students

• identify problems that lend themselves to a robot AI solution and decide whether a “classical” or data-driven solution should be preferred,
• cast an embodied-agent problem to a form that lends itself to an AI solution,
• generate an intelligent robot behavior (conceptual or in software):
  • Extract information from sensor streams (e.g., object/people identity/position, body postures, 3D room and object structures),
  • Learn useful sensorimotor behaviors (e.g., mobility or grasping), learn to analyse robotics applications from a system-level point of view, since robotics is very much a science of integration.
  • are stimulated to develop a critical, research-oriented attitude.

Prerequisites

This course is accessible as an optional course to last-year master students, or to master-after-master students. Hence, a master level background is expected, so that students can bring in real knowledge, in formal ways, from different background expertises, and combine it with that of other students.

The course concept (guided self-study) facilitates the entry of students with very different backgrounds. The number of students can be limited, because the system of guided self-study is very labour-intensive for the lecturer.

Content

This course is organized as guided self study: there is only a limited number of lectures in class (to explain and discuss the fundamental concepts of robot AI). For the rest of the course the students work on problems of their own choice. Collaboration in groups of maximally three students is encouraged.

The course has no organized examination session: it uses continuous evaluation, based on the students’ reports, to which feedback is provided by the lecturer and all other students. Reports and the feedback to them are public to all participating students, and become an inherent part of the “course material”. In a final individual discussion session with the lecturer, each student is expected to present the material in a relevant academic research paper in a very critical way, and to show creativity in identifying appropriate applications, open problems, or inherent limitations in the studied material.

The concept of the course allows to adapt its contents to the interests and background of the students.

Course Material

The core lectures of the course are based on the book Robotics, Vision and Control, written by Peter Corke, published by Springer in 2023. The e-book is free to download on Springer Link when connected to KU Leuven’s wifi or wired network.

In addition, students must find discussion material online, and contact the lecturer to select appropriate material of sufficient quality. The reports created by students, as well as the feedback that is provided to these reports, become an inherent part of the “course material”.

Evaluation

The course has no organized examination session: it uses continuous evaluation, from day 1 to the last day of the semester.

The grading is done based reports submitted during the year, and on one interactive and one-on-one discussion with the lecturer. The expected quality and content of the reports grow during the semester, along with the growing insights of the students in the material.

Reports are prepared “in the open”, via discussions in the lectures, and via a mailing list. Students can request dedicated lecture sessions, to discuss particular aspects that are not very clear, yet, or that raise particular interest for whatever reason (long as the contents has the focus of the course).