Sensitivity Analysis for Tractor-Trailer Dynamics – Prediction Model Quality Assessment

Institut

Lehrstuhl für Fahrzeugtechnik

Typ

Bachelorarbeit / Masterarbeit /

Inhalt

experimentell / theoretisch /  

Beschreibung

Trucks are one of the main carriers of goods. The trend of autonomous trucks offers many advantages for the freight transport sector including faster transport by avoiding mandatory breaks, solving the driver shortage problem, more traffic safety by anticipatory planning, and environmental benefits.

In the Chair of Automotive Technology, we are working with various cooperation partners from industry and other research institutes on the ATLAS-L4 project. Our main goal is to make truck-trailers drive hub-to-hub through highways autonomously (level 4).

Within the scope of autonomous driving, the longitudinal and lateral motion control of the truck to follow the planned trajectory is a challenging task. One of the common approaches for autonomous trajectory-following is Model Predictive Control (MPC). The goal is to precisely model the vehicle dynamics to minimize parameter uncertainties and reduce the influence of external disturbances such as crosswind, varying trailer load, or road banking, but still, fulfill the real-time execution requirements. To achieve a robust control performance, an identification/quantification of parameter uncertainties and disturbances is needed.

Besides improving the system state prediction, the results of a sensitivity analysis could help to deal with the trade-off between model complexity and real-time execution constraints.  

The following work packages are included in the bachelor’s thesis to be assigned:

• Familiarization with the fundamentals of the vehicle dynamics of a tractor-trailer system
• Identification of uncertainties and disturbances in the context of tractor-trailers and their effects (e.g. CG displacement)
• Research, selection, and implementation of suitable sensitivity analysis methods for the different uncertainties/disturbances based on chosen truck dynamics models (dynamic nonlinear single-/ double-track models).
• Identification of uncertainties/disturbances that affect the prediction quality the most (open-loop prediction error).
• Extension of the models by modeling the most relevant uncertainties and disturbances and evaluation of the prediction quality of the extended ones
• Writing a thesis report

The thesis should document the individual work steps in a clear form. The candidate undertakes to complete the bachelor’s thesis independently and to indicate the scientific resources used. The submitted thesis remains the property of the chair as an examination document.

Voraussetzungen

• Motivation to familiarize yourself with new topics and to try new ideas
• Ideally previous experience with Simulink/Python/C++, Git, ROS
• Ideally previous knowledge in vehicle dynamics or control theory

Tags

FTM Studienarbeit, FTM Automatisiertes Fahren, FTM IVS, FTM Zarrouki, FTM Informatik

Möglicher Beginn

sofort

Kontakt

Baha Zarrouki, M.Sc.
Raum: MW3527
Tel.: +49 (89) 289 - 10498
baha.zarroukitum.de

Ausschreibung