Motion control of autonomous vehicles with stochastic nonlinear Model Predictive Control
- Lehrstuhl für Fahrzeugtechnik
- Semesterarbeit Masterarbeit
- experimentell theoretisch
Within the scope of fully autonomous driving, controlling the vehicle’s lateral- and longitudinal dynamics to follow a planned trajectory is a challenging task.
One of the common approaches for autonomous trajectory following is Model Predictive Control (MPC). MPC determines the optimal steering and/or acceleration inputs for each time step by solving an optimization problem that is specified by a cost function, which formulates our control objectives, and constraints, that consider the vehicle limitations and safety aspects. Based on predicting the vehicle’s future states over a certain horizon, the MPC can act and plan in advance the optimal way to follow the reference trajectory computed by the motion planner.
Nominal MPC formulations do not consider parametric uncertainty in the prediction model and disturbances, tending to make optimistic predictions leading to suboptimal solutions . Robust MPC methods, e.g. using constraint tightening techniques, ensure constraint satisfaction under certain disturbances. However, they show a very conservative performance. In contrast, stochastic MPC schemes [1-2] do not consider hard constraints but promise to satisfy constraints with a certain probability making them less conservative.
In this thesis, the problem of the combined longitudinal- and lateral control of an autonomous vehicle for trajectory following is addressed with a stochastic nonlinear MPC approach.
The following work packages are included in the thesis to be assigned:
- Literature research
- Familiarization with the available TUM Control Simulation Framework and extension with needed features
- Development of a suitable stochastic nonlinear MPC for trajectory following
- Evaluation of the controller performance and robustness
- Writing a scientific thesis report
The thesis should document the individual work steps in a clear form. The candidate undertakes to independently complete the bachelor’s thesis and indicate the scientific resources used.
The submitted thesis remains the property of the chair as an examination document.
 Liniger, Alexander, et al. "Racing miniature cars: Enhancing performance using stochastic MPC and disturbance feedback." 2017 American Control Conference (ACC). IEEE, 2017.
 Rawlings, James Blake, David Q. Mayne, and Moritz Diehl. Model predictive control: theory, computation, and design. Vol. 2. Madison, WI: Nob Hill Publishing, 2017.
- Motivation to familiarize yourself with new topics and to try new ideas
- Ideally previous theoretical knowledge in Model Predictive Control
- Ideally previous experience with Python/C++, Git, ROS2
- FTM Studienarbeit, FTM AV, FTM Zarrouki, FTM Informatik, FTM AV Safe Operation
- Möglicher Beginn
Baha Zarrouki, M.Sc.
Tel.: +49 (89) 289 - 10498