0000002736 20W 6SWS PR Masterpraktikum - Intelligent Mobile Robots with ROS   Hilfe Logo

LV - Detailansicht

Wichtigste Meldungen anzeigenMeldungsfenster schließen
Allgemeine Angaben
Masterpraktikum - Intelligent Mobile Robots with ROS 
practical training
Winter semester 2020/21
Informatics 6 - Chair of Robotics, Artificial Intelligence and Real-time Systems (Prof. Knoll)
(Contact information)
Allocations: 1 
Angaben zur Abhaltung
Important: There will be an online kick-off meeting held on July 14th at 4 pm using zoom. Please attend this meeting if you are interested in this course.

Meeting Link: https://tum-conf.zoom.us/j/92083728698
Meeting ID: 920 8372 8698
Password: 826135

Autonomous mobile robots have been a research area of interest for decades now. They have come a long way from first navigation approaches using large, single-core computers and ultra-sound sensors to fully autonomous machines equipped with GPUs and advances sensors like 3D lidars. Nowadays, they are frequently used in logistic centers, healthcare system or department stores. They are able to operate safely and autonomously in their environment as well as detect and interact with people.
In this course, students will get the chance to understand all parts of a mobile robot (software & hardware) and work on their own, self-chosen project on a real mobile robot (Robotino) equipped with multiple different sensors.

This course attempts to give a practically oriented overview of all disciplines with the field of mobile robotics. This includes the localization, mapping, navigation and perception of the robot in an (un-)known environment. In order to master this knowledge, the first part of the course will be consisting of (online) lectures, each focusing on one of the previously mentioned disciplines. At the end of each lecture the student has to work on a small practical exercise in order to apply the theoretical knowledge in the real (simulated) world.
During the second part of the course, students will form heterogeneous teams of 3-5 members, preferably with different backgrounds. Each team will then choose a problem or task to work on freely. This task could be almost anything, from developing a delivery robot (more software oriented) to trying to solve a known problem (e.g. localization) with a new algorithm (e.g. spiking neural networks) (more research oriented). This allows the students to choose a topic aligned with their personal interests and knowledge.

Due to Corona restrictions, the first part of the course will most likely be held completely virtual, including virtual lectures and exercises using a simulator. The second part, however, will definitely take place in the lab at Hochbrück where you can pick one of multiple Robotino robots for your final project.

Since the final project is team based, this course is suitable for students with almost any kind of background. As long as you are interested in mobile robots and motivated to work in a team on an actual robot to solve a problem, you are going to fit right in. (Of course you need to have some kind of programming experience, see next paragraph)
To participate in this course successfully, practical programming skills in C/C++ or Python are mandatory as well as good knowledge of Linux. First experiences with the Robot Operating System (ROS) are beneficial. Preferably, the students also have prior knowledge in at least one of the following fields: localization, mapping, perception, navigation, control, deep learning, reinforcement learning, software engineering.
After successful participation in this course, students will be able to

  • work with a mobile robot in simulation as well as the real-world

  • identify and understand the components a mobile robot requires for autonomous operation

  • work as a "specialist" a heterogeneous team composed of members with different backgrounds

  • pick a topic, define it and develop a structured plan to work on it

Beyond that students will have aquired the following competences:

  • basic understanding of the Robot Operating System (ROS) including programming experience in either C++ or Python

  • knowledge of current algorithms in localization, mapping and control (perception)

  • English
  • German
The first part of the course will be most likely held virtually using zoom/online lectures. The second part of the course, the project, will be taking place at the lab at Parkring 13 in Garching.
Für die Anmeldung zur Teilnahme müssen Sie sich in TUMonline als Studierende/r identifizieren.
Note: For registration you have to be identified in TUMonline as a student.

Note: The registration for the course will be managed by the TUM matching system. Applicants are required to send a brief description of their skills and practical programming experience together with a short paragraph outlining their motivation to join the course to robin.dietrich@tum.de. Please specify the subject of the email according to the following template:

[course#] Application: first name last name

e.g. for this course and a student named Manfred Müller

[0000002736] Application: Manfred Müller
If you are already interested in the field of intelligent mobile robots and want to gain some knowledge prior to the courses starting date the following literature, lectures can get you started. Some of them are only listed in case you plan on focusing your final project on that (e.g. deep, reinforcement learning).


  • Probabilistic Robotics by Thrun et al. (the number one book for localization, mapping and planning of mobile robots)

  • Multiagent Systems by Weiss (covering basics and advanced theories in communication and cooperation of multiple agents in a single system)

  • Deep Learning by Goodfellow et al. (good starter into deep learning in case you want to do something in that direction for your final project)

Online Lectures:
Online information
e-learning course (moodle)