Data-Driven Algorithms in Vehicle Technology

  • Typ: Vorlesung
  • Semester: Wintersemester
  • Ort:

    Campus Ost, Geb. 70.04, Raum 219

  • Zeit:

    Wintersemester

  • Beginn: 24.10.2023
  • Dozent:

    Dr.- Ing. Stefan Scheubner

  • ECTS: 4
  • LVNr.: 2113840
  • Hinweis:

    Die Einführungsveranstaltung findet am 24.10.2023 um 14:00 Uhr in Präsenz am Campus Ost, Geb.70.04, Raum 219 statt. Alle weiteren Vorlesungsinhalte werden als Videoaufzeichnungen in ILIAS bereit gestellt. In regelmäßigen Abständen wird es Sprechstunden geben. Die genauen Termine erfahren Sie dann über den entsprechenden ILIAS Kurs.

Course Syllabus: Data-Driven Algorithms in Vehicle Technology

Motivation for the Course: Nowadays, engineers often develop technical systems using a combination of hard- and software. This is true especially for modern passenger vehicle development. In a digitalized world, such developments are built on knowledge gained from relevant data sources, e.g. the vehicle sensors. Therefore, engineers in automobile technology need qualifications from data science to successfully create new functionalities in the cars. To prevent remaining purely theoretical, the algorithms in this course are explained using a real-world problem of “EV Routing”. Students have the opportunity to test methods in Python with frequent exercises presented.

Goal of the Course: Students have a basic understanding of data-driven algorithms such as Markov Models, Machine Learning or Monte-Carlo Methods. The approach for building data-driven models in automobile technology are known to students and they are able to test algorithms in the programming language “Python”. Furthermore, students have learnt how to analyse the algorithm performance.

Content:

1. Introduction to function development as well as the prerequisites for the course (e.g. Fundamentals for running Python code)

2. Fundamentals for EV Routing and relevant data sources

3. Parameter estimation and state classification algorithms to determine the current situation of the vehicle

4. Learning methods for driver behaviour

5. Forecast algorithms to predict future energy consumption of an electric vehicle