Paper presented at the 2nd international conference on driver behaviour and training, held Edinburgh, November 2005, and also published in Driver Behaviour and Training volume 2

Vehicular traffic flow is a complex interaction between drivers, vehicles, and the road infrastructure. These interactions can be viewed as human-machine systems, describing the relationship between driver (human) and vehicle (machine) and the reciprocity of the driver-vehicle system with the environment. The driver component of the system is the most complex as they are generally characterised by higher-level processes. These processes include perceptual capabilities such as vision, hearing and sensation of forces on the body; cognitive functions, for example motivation and attitude; and control functions, such as limb coordination enabling steering and braking. Existing methodologies for representing these processes are limited in three fundamental ways: (1) the ‘lower level’ of detail in the modelling (2) the lack of behavioural intelligence within the modelled system and (3) the failure to integrate the three key elements of driver, vehicle and road infrastructure. The paper examines various driver behaviour models. It then focuses on the development of a framework for complex and interactive scenario modelling, particularly applied to hazard perception for driver training. A key element of this framework is the flexibility of qualitative visualisation, enabling a scenario to be seen from various viewpoints. The main aim of developing such a framework is to provide a modelling environment that can integrate theory, algorithms, software and experimental results generated by engineers and scientists in all of the traditionally disparate areas of driver-, vehicle-, traffic- and highway research. The framework has been validated using empirical data obtained from TRL’s driving (car) simulator, to assess driver perceptual and cognitive skills required in vehicle control to avoid collision with a parked car. In the context of hazard perception, the validation has demonstrated the potential of using the framework as a comprehensive modelling tool for engineers, scientists and decision-makers working in many commercial aspects of road transport.

Want to know more about this project?