Modellering van bestuurdersgedrag en gedragsaanpassing aan intelligente voertuigen in een stedelijke omgeving

Modelling of driving behaviour and behavioural adaptation to advanced driver assistance systems (ADAS) when driving on urban intersections


Drs. Nina Schaap


prof. dr. ir. Bart van Arem


July 2005 – April 2010

The driving task is a hierarchical task with three interacting levels. This interaction becomes most clear on urban intersections, where all levels are equally important. A behaviour simulation model of intersection driving will be developed, based on this hierarchical structure of the driving task. With this driver model, behavioural adaptation to ADAS on urban intersections can be studied. A first explorative experiment focussed on the reactions to unexpected events; next steps are to study the aspects of secondary tasks and behavioural characteristics which might influence these reactions to unexpected events.


ADAS, behavioural adaptation, driver model, urban intersection, workload, unexpected event

Background and problem definition

The three interacting levels of the driving taskThe effects of behavioural adaptation to ADAS for urban intersections are difficult to study before their introduction. Therefore, a driving behaviour simulation model will be developed, which will be able to interact with ADAS. By using the structure of the driving task, it becomes possible to develop a intersection driving model which is extended beyond single subtasks.

The driving task is a hierarchical task with three interacting levels. On the strategic level, the goals of the trip and the route are set. This determines the manoeuvres the driver has to make on the tactical level, which then determines the way the vehicle has to be controlled on the operational level (top-down influence). Alternatively, actions and decisions on a lower level can influence goals and priorities of higher level tasks in unexpected situations (bottom-up influence). This happens when the drivers encounters an unexpected situation: the driver compensates and changes his behaviour temporarily.

TNO driving simulatorMethodological approach

The driver model will be based on literature research and on two driving simulator experiments. The first experiment focussed on the interaction between levels of the driving task and on situations in which this interaction is most apparent. This experiment showed that unexpected events invoke temporary behavioural changes on the tactical level. The effect of workload and the level of seriousness of the unexpected events was also studied.

Further experiments will look further into how workload influences normal driving behaviour and the reactions to unexpected events.

Expected practical and scientific results

A behaviour simulation model of intersection driving will be developed. This will result in new knowledge about the structure of the highly complex driving task and the underlying processes of behavioural adaptation to ADAS. Furthermore, new insight into how workload affects the reactions to unexpected events will be gained.

Scientific and societal relevance

The behaviour simulation model will give new insight into driving behaviour and the driving task. It can also be implemented into the driver module of simulation software (e.g., driving simulator software and microscopic traffic simulation models) to improve its performance. New knowledge about behavioural adaptation to intersection ADAS can be used for further development of new ADA systems.

Selected publications

Schaap, T.W., A.R.A. van der Horst & B. van Arem (2008) Influence of unexpected events on driving behaviour at different hierarchical levels: a driving simulator experiment, Proceedings: European Conference on Human Design for Intelligent Transport Systems, 3-4 April 2008, Lyon [PDF]

Schaap, T.W. & B. van Arem (2006), A comprehensive driver behavior model for the evaluation of intelligent intersections, In Proceedings ITS World Congress 8-12 October 2006, London, Paper 1305

Schaap, T.W. & B. van Arem (2006), An evaluation of behaviour simulation modelling tools for urban intersection driving, TRAIL Congress, 21 November 2006, Rotterdam, The Netherlands