227-0085-24L  Projekte & Seminare: Vision and Control in RoboCup

SemesterHerbstsemester 2022
DozierendeJ. Lygeros, L. Van Gool, F. Yu
Periodizitätjedes Semester wiederkehrende Veranstaltung
LehrspracheEnglisch
KommentarNur für Elektrotechnik und Informationstechnologie BSc.

Die Lerneinheit kann nur einmal belegt werden. Eine wiederholte Belegung in einem späteren Semester ist nicht anrechenbar.


KurzbeschreibungDer Bereich Praktika, Projekte, Seminare umfasst Lehrveranstaltungen in unterschiedlichen Formaten zum Erwerb von praktischen Kenntnissen und Fertigkeiten. Ausserdem soll selbstständiges Experimentieren und Gestalten gefördert, exploratives Lernen ermöglicht und die Methodik von Projektarbeiten vermittelt werden.
LernzielVision and Control in RoboCup is jointly offered by Prof. John Lygeros (IFA), Prof. Luc Van Gool (CVL) and Prof. Fisher Yu (CVL).

RoboCup is a tournament where teams of autonomous robots compete in soccer matches against each other. The ETH team NomadZ ( Link ) plays in the Standard Platform League with a team of humanoid NAO robots, where the focus lies on developing robust and efficient algorithms for vision, control and behavior.
The main objective of this course is to familiarize students on the fundamental challenges we encounter in RoboCup. This is accomplished by a combination of lecture sessions, related student exercise sets and programming projects in MATLAB and Python. The topics cover visual localization, deep learning for object detection and reinforcement learning control of unknown systems.

Important information for candidates:
You are required to bring your own Laptop for the programing exercises.

The course is taught in English and open to 5th or higher semester students.
A basic knowledge of programming in Python and MATLAB is required.
A prior exposure to control theory (e.g. by attending a Control Systems course) is desirable. Those students who are not familiar with control theory will need to complete some extra study to understand some aspects of this P&S.