Model Predictive Control & Machine Learning for Autonomous Vehicles

Institut
Lehrstuhl für Fahrzeugtechnik
Typ
Bachelorarbeit / Semesterarbeit / Masterarbeit /
Inhalt
experimentell / theoretisch /  
Beschreibung

At the Chair of Automotive Technology, we are developing a new full open-source software stack to operate autonomous vehicles not only in a motorsport context, but also on public roads: the AV2.0. The AV 2.0 represents a scalable and modular platform. The main modules are: Perception, Planning and Control. This enables us to quickly evaluate novel concepts and to operate our different experimental vehicle types, e.g. multivan, race car or truck.

Do you want to put your own mark on the future of autonomous vehicles with your ideas and concepts? You will be part of a team conducting state-of-the-art research in the trajectory control of autonomous vehicles.
If you are interested in a student research project, feel free to send me an initiative application and we will arrange a call to discuss the topics. Just send me an e-mail with a short motivation, curriculum vitae, and a recent transcript of records.

Currently, the following topics can be addressed:

  • Vehicle Dynamics Modeling:
    • Development and implementation of a non-linear truck dynamics model in the high-velocity range.
    • Approximation of the model for real-time feasibility, e.g. using (cascaded) neural networks and investigation of different approximation strategies, e.g. end-to-end- or modular cascaded approximation.
    • Online learning/ online parameter identification for an adaptive dynamics model
  • Robust Vehicle Dynamics Control:
    • Investigation and modeling of uncertainties and disturbance in the vehicle context (e.g. crosswind)
    • Sensitivity analysis of the uncertainties and disturbances on the system state.
  • Model Predictive Control (MPC):
    • Design of a robust real-time capable non-linear MPC for trajectory tracking
    • Learning-Based MPC
    • Adaptive robust MPC
    • Neural Network-Based MPC (NNMPC)
    • Reinforcement Learning of MPC parameters

The student research project/ thesis will handle upon agreement a single or multiple bullet point work packages.   

Voraussetzungen
  • Motivation to familiarize yourself with new topics and to try new ideas
  • Ideally previous experience with Simulink/Python/C++, Git, ROS2
  • Ideally previous knowledge in vehicle dynamics or control theory
Tags
FTM Studienarbeit, FTM AV, FTM Zarrouki, FTM Informatik, FTM AV Safe Operation
Möglicher Beginn
sofort
Kontakt
Baha Zarrouki, M.Sc.
Raum: MW3505
Tel.: +49 (89) 289 - 10498
baha.zarroukitum.de
Ausschreibung