Deep Robot Learning


SS 2017





7 ECTS credits


Note: This page is for a course from a previous semester.
Find a list of current courses on the Teaching page.

Course Description:

Deep learning currently has tremendous impact on the fields of computer vision and robotics. In this practical course, students will implement and apply deep learning algorithms for robotics tasks such as vision-based navigation for quadrotors or vision-guided object manipulation in a robot simulation environment. Depending on the available project tasks, working with an actual robot might also be possible.

In a first phase (4 weeks), there will be lectures and practical exercises on a weekly basis. In a second phase, the students will be assigned a practical project and work in teams of 2 students on the project. This second phase lasts for the remainder of the lecture period. Each group meets weekly with their tutor and presents and discusses its progress (meetings mandatory!). At the end of the course, the teams will present their project outcome in a talk and demonstrate their solutions. They will document their project work in a written report.

The course will take place in seminar room 024 (lectures) and room 125 (exercises and project) in the UMIC building (Aachen).

Course Registration:

Application for registration only through the Seminar and Practical Project Seminar (Praktikum) Registration. When you apply for registration, please also write in the respective text field of the application form how you fulfill the course requirements (see below). The application process is now closed.

Max no. of participants: 8

Course Layout:

  • Lecture & Exercise (tba): 2 hours per week lecture session, Wednesdays from 2pm to 4pm (introductory lecture on April 19th) in room 024, UMIC building. 2 hours per week tutored exercises, Wednesdays from 4pm to 6pm. There are 4 lecture & exercise sessions. Each week, the exercise for the following week will be announced and the students present their solution for the exercise of the current week to the tutor. The exercises must be done in groups of 2 students. The groups should be formed on the first introductory lecture day. Students can use lab computers in room 125 in the UMIC building. Attendance to lectures and exercises is mandatory.
  • Project (tba): Each group will be assigned to a project after the lecture & exercise phase. Students can work on the project in the lab and consult the tutor on Wednesdays from 2pm to 6pm. Attendance to weekly meetings with tutor is mandatory. Additional lab time for working freely can be arranged.
  • Presentation and demo: Each group will be assigned a time slot on one of the last days of the semester lecture period, to present their results and give a live demo, followed by a Q&A session. Date shortly after the lecture period (Wed, August 9th, 2017, 2pm-6pm).
  • Project Report: Each group writes a report on their project work (12 pages, single column, single-spaced lines, 11pt font size). Due date shortly after the lecture period (due date Fri, August 18th, 2017).

Course Requirements:

  • Good knowledge of python, C/C++ and basic mathematics such as linear algebra, analysis and numerics is required
  • Prior practical knowledge in deep learning frameworks (e.g. theano, torch, tensorflow) and robotics/computer vision topics is a plus
  • Participation in at least one of the following lectures of the RWTH Aachen Computer Vision Group: Machine Learning, Advanced Machine Learning. Similar lectures that treat deep learning can also be accepted, please contact us.

Course Materials:

tba in the introductory lecture.

Course Schedule
Date Title Content Material
Introductory Lecture Organizational issues, introduction to deep robot learning, deep learning basics, introduction to TensorFlow
Exercises Deep learning basics and TensorFlow
Lecture Reinforcement learning
Exercises Reinforcement learning
Lecture Deep reinforcement learning
Exercises Deep reinforcement learning
Lecture Introduction to ROS
Exercises ROS
Lab Introductory meeting for project phase, tutored lab time
Lab Tutored lab time
Lab Tutored lab time
Lab Tutored lab time
Lab Tutored lab time
Lab Tutored lab time
Lab Tutored lab time
Lab Tutored lab time
Final Presentation & Demo Final presentation and demo in the lab room
Final Report Due date final report
Disclaimer Home Visual Computing institute RWTH Aachen University