Community Research and Development Information Service - CORDIS

Method for evaluation of the imnpact of intelligent road transport driver systems on driver behaviour

The problematic nature of assessing the impact of in vehicle transport systems (ITS) technology is rooted in the highly complex and integrated nature of the driving task itself. Driving may be described as an act of dynamic control which relies on highly proceduralised knowledge. This activity is not only dynamic but is probabilistic, the driver must be proactive and selective when seeking out relevant task information from the traffic environment. The problem for the project was twofold. Firstly, ITS systems may affect driving behaviour through any number of complex and subtle ways which are difficult to capture as objective, hard data. Secondly, techniques of data capture must be sufficiently sensitive to pinpoint those devices which could distract or reduce the capacity of the driver to safely control the vehicle.
The development of a conceptual framework describing driving behaviour was central to directing the technical aims of the project. The important aspect of the framework is the acknowledgement that driving is a primarily cognitive as opposed to a motor activity. Therefore usability may not be assessed solely within the simplistic terms of a stimulus response model (ie simply install a ITS device and observe overt changes in driving behaviour). It is essential to consider the more nebulous effects of technology on cognitive aspects of the driving task such as decision making and perception (ie at the strategic level). This covert category of data may only be captured via psychological techniques. Single technique approaches are liable to produce inaccurate or misleading conclusions. The strength of the multilevel approach is that it allows certain types of data to be cross-referenced to those obtained via other techniques.
A technical strategy of the project was to develop and validate a set of data capture techniques on the basis of the above framework through a process of test retest validity. In addition the project aimed to address the question of dat a capture environments (ie where to perform usability evaluations). This issue was examined via 2 parallel strands of experimentation, one to be carried out in a dynamic vehicle simulator and the other using instrumented vehicles on the real road.
Irrespective of data capture environments, the project intended to demonstrate via application how the data tools should be used as part of a usability analysis. Usability evaluation method and information were directed towards changes in driver behaviour at 4 levels: group perception and strategies (eg cultural norms), individual perceptions, attitudes and strategies, individual control behaviour and individual physiological responses. The first 2 levels produce qualitative, descriptive data which addresses issues of system acceptability. Whereas the next levels attempt to quantify those changes in driver behaviour in terms of ITS system safety and efficiency.
The project has developed a body of knowledge, a level of expertise and documented research facilities that will have a usefulness beyond the life of this particular project.

Reported by

Loughborough University
The Elms Elms Grove
LE11 1RG Loughborough
United Kingdom
See on map
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top