UR:BAN - Prediction of Driver Behaviour

Below you will find a short summary of the project. For further information please contact the project managers via the contact button on the right side.


The project “UR: BAN” (Urban Space: User-Friendly Assistance Systems and Network Management), funded by the Federal Ministry for Economic Affairs and Energy, deals with a wide range of issues concerning driver assistance systems for city traffic. The project involves a total of 30 partners from the automotive and supply industries, electronics, communications and software companies, universities, research institutes and cities.

The project is divided into 3 major subprojects:

  • Cognitive Assistance
  • Networked Transport System
  • Human in Traffic

The Institute of Ergonomics & Human Factors (IAD) is involved in the project as a subcontractor of the Opel Automobile GmbH in the subproject “Behaviour Prediction and Intention Recognition (VIE)”, which is part of the subproject “Human in traffic”.


The subproject “Human in traffic” deals with the question of how the driver can travel safely, efficiently and comfortably in urban traffic. Within the framework of the cooperation between the IAD and the Opel Automobile GmbH, it is examined to which extent the prediction of the driver's future actions can help to improve the effect and the acceptance of a driver assistance system.

On the one hand, it is essential for an assistance system to be effective at a very early stage. On the other hand, however, too early warnings or interventions often disturb the driver, especially if he has control of the emerging situation and the warning for his upcoming reaction is no longer appropriate. This so-called “warning dilemma” can lead to a significant decrease in the acceptance of the system.


The driving tests for this project were carried out on the test area of the TU Darmstadt. A total of 102 subjects participated in two test series in the summers of 2013 and 2014, in which the following driving manoeuvres were provoked by different experimental equipment:

  • lane change manoeuvre
  • stopping manoeuvres
  • emergency evasion manoeuvres
  • emergency braking

In addition to so-called CAN bus data (e.g. the current steering wheel angle or the pedal positions), eye movement data and data from various questionnaires were collected in the tests and subsequently analysed.


An algorithm (approach of a fuzzy logic) was developed on the basis of this data, which made it possible to predict the behaviour of the drivers with a time horizon of approx. 1-2s. The algorithm was presented at the closing event in autumn 2015 in Düsseldorf.


Heine, J. (2017). Entwicklung eines Algorithmus zu Prädikation eines innerstädtischen Fahrstreifenwechsels. [Dissertation] Technische Universität Darmstadt.

Langer, I. (2016). Analyse von Aktivitäten eines Fahrzeugführers zur Verhaltensbeschreibung am Beispiel des Fahrstreifenwechsels. [Dissertation] Technische Universität Darmstadt.

Langer, I., Holzheimer, F., Heine, J., Abendroth, B., & Bruder, R. (2015). Development and partial validation of a catalogue of action steps for a car driver. In: Proceedings 19th Triennial Congress of the IEA, 09.-14. August 2015, Melbourne, Australien.

Heine, J., Sylla M., Langer, I., Schramm, T., Abendroth, B., & Bruder, R. (2015). Algorithm for driver intention detection with Fuzzy Logic and Edit Distance. In: IEEE 18th International Conference on Intelligent Transportation Systems ITSC, 15.-18. September 2015, Las Palmas de Gran Canaria, Spanien.

Heine, J., Krämer I., Achieser, I., Langer, I., Schramm, T., & Abendroth, B. (2015). Bewertung von Prädiktoren zur Fahrerintentionserkennung. In: 5. Berliner Fachtagung Fahrermodellierung, 11. Juni 2015, Berlin.

Langer, I., Heine, J., Abendroth, B., & Bruder, R. (2014). Herausforderungen beim Hervorrufen von kritischen Brems- oder Lenkreaktionen in Fahrversuchen zur Untersuchung der Fahrerintention. In: 60. GfA-Frühjahrskongress, 12.-14. März 2014, München.

Langer, I., Heine, J., Schramm, T., & Bruder, R. (2013). Kritikalitätsmaß einer Fahrsituation – Eingangsgröße für einen Algorithmus zur Fahrerintentionserkennung. In: VDI Berichte 2205 – Der Fahrer im 21. Jahrhundert. Düsseldorf: VDI Verlag GmbH, S. 75-99.