smart TUBe - Team of TU Berlin at competitions of autonomous driving scaled vehicles
The smartTUBE team is a student team that is continuously developing a model vehicle for competitions of autonomously driving scaled vehicles (e.g. Carolo Cup or VDI Autonomous Driving Challenge) in the project "Automated Driving". For this purpose, the Department of Motor Vehicles provides premises in TIB 13, a test track as well as software and technical equipment.
The team consists of 10 to 15 students from different courses of study.
The competition for autonomous driving model cars on a scale of 1:8 will take place again next year, probably from May 24 to 26 of May 2024, as part of the DTM race (https://dtm.com/residencial/events/lausitzring-2024) at the DEKRA Lausitzring.
The vehicle and the technology for autonomous driving (https://mdynamix.de/mx-products/mx-carkits/) will be provided to us by our cooperation partner "DEKRA Lausitzrring" (https://www.dekra-lausitzring.de).
Task
In competitions of autonomously driving scaled vehicles (e.g. Carolo Cup), the aim is usually for student teams to compete within a competition in the field of autonomous driving. This gives students the opportunity to deal with the topic of automated driving in a practical way and to gain experience.
Various disciplines must be completed for successful participation. In addition to a presentation of the overall concept, the autonomous driving functions are evaluated. For this purpose, a circuit similar to the available test track must be driven both with and without obstacles. In addition, the vehicle has to detect a parking space independently and park in it within a short time.
Team
The project's premises are available throughout for working on the tasks. In addition, the team meets weekly to discuss work progress and technical problems. In addition, the team is supervised by Ingo Wülfing.
The tasks worked on by the team members are assigned to different modules. These cover the topics of construction, electronics, controller design, image processing and sensor evaluation. The results are presented in four presentations during the semester.
Hardware and software architecture
The vehicle is enhanced with sensors, microcontrollers and a computer. A camera as well as lidar and ultrasonic sensors are used for environment detection. The actuators are controlled and the sensor data is processed on an NVIDIA Jetson Xavier NX. As software environment ROS (Robot Operating System) (http://wiki.ros.org) and Python (https://www.python.org) are used.
The development environment is "Visual Studio Code" (https://code.visualstudio.com), GIT (https://git.tu-berlin.de/FGKfz-smartTUBe/vdi-cup/) and Matlab/Simulink (https://de.mathworks.com/products/matlab.html).
The exact MXcarkit specifications:
https://mdynamix.de/mx-products/mx-carkits/mxcarkit-spezifikationen/
https://github.com/william-mx
