Thursday, April 24, 2025

Semi-automated labeling of real driving data

Topic and Goal of the Thesis

The driving profile is an essential input for the simulation of motor vehicles. There are synthetic driving profiles such as the WLTC, but real journeys can also be used for this purpose. To compile scenarios from real driving data, the data must be labeled so that the correct cycles can be retrieved depending on the desired label. In this work, this classification is to be carried out semi-automatically.

Working Points

In the course of this work, a methodology is to be developed with which an existing data pool of real driving data is used to derive labeled cycle data.

• Research on cycle derivation from real driving data
• Development of a process for cycle derivation and classification
• Implementation of the process in Matlab and Python
• Creation of driving scenarios using the classified data

Requirements

• Good English or German language skills
• Reliability, commitment and enjoyment of working independently

Note: Please attach brief resume and grade summary.