Assistance Systems to Recognize the Behavioral Intention of other Road Users

Recognizing the intentions of drivers and factoring in the wishes and intentions of users is a difficult but important aspect of user-oriented system design. Whenever a human-computer, human-robot or human-vehicle interaction has the human side throwing their hands up in the air and asking, “What’s it doing now?” – that is a sign that these systems do not behave as users expect them to.

ADAS Behavior Prediction

Video “Driver-vehicle interaction” (example)
If a driver intends to pass another car and thus reduces the distance to it, adaptive cruise control should recognize this intention and not brake.

Only when a driver intends to make a turn at an intersection should they be warned of a cyclist or pedestrian who is moving parallel and about cross the road the car is about to turn into. If drivers were warned of every cyclist or pedestrian near their vehicle, acceptance of the system would be undermined and drivers would start looking for the off switch. One particular challenge in this situation is to predict the behavior of pedestrians as they can very quickly change their direction and speed.

The objective of developing systems that predict behavior is to rely on “standard” sensors as much as possible and ensure a high hit rate with few false alerts – and all of this in real time.

ADAS Behavior Prediction (video in German)

Video “Behavior prediction – recognizing intent”
The project UR:BAN has devoted a specific subproject to this issue.

Verhalten_1.jpg   Verhalten_2.jpg

Traffic not visible                                                 Feet not visible

Verhalten_3.jpg   Verhalten_4.jpg

Heads not visible                                                Only position visible





Graichen, Matthias: Analyse des Fahrverhaltens bei der Annäherung an Knotenpunkte und personenspezifische Vorhersage von Abbiegemanövern. Dissertation Universität der Bundeswehr München, 2019. 

Graichen, Matthias; Nitsch, Verena: Effects of driver characteristics and driver state on predicting turning maneuvers in urban areas. Is there a need for individualized parametrization? In: Stanton, Nevilla A.; Landry, Steven; Di Bucchianico, Giuseppe; Vallicelli, Andrea (Eds.) Advances in Human Aspects of Transportation, 2017. p 15-29. Proceedings of the AHFE 2016 International Conference on Human Factors in Transportation, July 27–31, 2016.

Graichen, Matthias; Nitsch, Verena; Färber, Berthold: A Meta-perspective on Research Activities in UR:BAN Human Factors in Traffic. In: Bengler, Klaus; Drüke, Julia; Hoffmann, Silja; Manstetten, Dietrich; Neukum, Alexandra (Eds.): UR:BAN Human Factors in Traffic – Approaches for Safe, Efficient and Stress-free Urban Traffic. ATZ-MTZ-Fachbuch. Springer Vieweg Wiesbaden, 2018, Pp.29-46.

Färber, Berthold: Kommunikationsprobleme zwischen autonomen Fahrzeugen und menschlichen Fahrern. In: Maurer, Markus; Gerdes, J. Christian; Lenz, Barbara; Winner, Hermann (Hrsg.): Autonomes Fahren – Technische, rechtliche und gesellschaftliche Aspekte. Springer Vieweg Berlin, 2015, S.127-146. (kostenlos abrufbar unter:

Schmidt, Sabrina; Färber, Berthold: Pedestrians at the kerb – Recognising the action intentions of humans. In: Transportation Research, Part F, Traffic, Psychology and Behaviour, 12(4), pp. 300-310, July 2009.