Autonomous Drone Racing Project Course

Projektpraktikum

In this course, teams of up to 2 students will work jointly to solve the autonomous drone racing problem. The main goal of this course is to teach robotics problem-solving skills as well as project management and teamwork. This course was formerly offered as the "Robot Learning and Control Project Course."

 

Enrollment and Acceptance Procedure
To participate in the course, students must first register via TUMonline. Please note that it is standard procedure for all students to initially be placed on the waiting list upon registration. At the beginning of the semester, an initial qualification challenge will be conducted (further details will be provided during the first lecture). Teams of up to two students who successfully pass this challenge will be officially accepted into the course and moved off the waiting list. The course capacity is limited to 20 students; in the event that the number of successful applicants exceeds this limit, admission will be granted to the fastest teams based on their performance in the qualification challenge.

Credit Eligibility and Alternative Enrollment
Depending on your specific study program, you may not be eligible to receive academic credits for the standard Autonomous Drone Racing Project Course. To accommodate this, we offer the Multi-Agent Autonomous Drone Racing (MAADR) track, which is conducted as part of the Master Practical Course: Learning for Robotics on Real Hardware. If your study program does not grant credits for the regular course, please check your matching platform. If the Master Practical Course: Learning for Robotics on Real Hardware (IN2106) is listed there, you can enroll through that platform to participate in the MAADR track.

Learning Objectives

Students will learn to independently solve robotics control and learning problems. This includes analyzing the problem, developing a project plan, trading off different solution approaches, and demonstrating the results in a realistic simulation or on real robot hardware. By developing an idea into a real-robot demonstration, students will gain hands-on implementation experience in robot decision-making algorithms. Students will also learn to work efficiently as a team, document approach, and communicate results. The skills acquired in this course are preparation for research, entrepreneurship, or a leadership position in the industry. 

Prerequisites

The prerequisites for this course are fundamental knowledge in robotics, machine learning, control theory, and computer vision as well as experience in Python or C++ programming and Linux (Ubuntu). Experience in the Robot Operating System (ROS) is a plus.

Teaching and Learning Approach

Each team will work on an individual robotics problem with a dedicated advisor. Introductory lectures will be provided to teach approaches and tools for successful robotics problem-solving.  The advisors will support their teams in the process of finding relevant literature, refining the project, and evaluating different solution approaches. They will also give early feedback on the robot demonstration, report, and presentation.

Evaluation

The evaluation for this course is based on (i) the quality of the implementation and the results of the final robot demonstration, (ii) a written report summarizing the results, and (iii) a final project presentation followed by a Q&A session.