Institute for Transport Studies (ITS)

Developing next generation driving behaviour models

Supervisor: Dr Charisma Choudhury

Traffic simulation tools are widely used around the world for transportation planning and management. In particular, microscopic traffic simulation tools which replicate individual driver decisions and aggregate them to deduce network conditions are increasingly popular for the evaluation of alternative transport policies. An essential component of such tools is a set of mathematical models of driver behaviour, including but not limited to longitudinal movement models, lateral movement models, and route choice models. The development and calibration of such models rely on an in-depth appreciation of the complexity of driver behaviour.

While second nature to the majority of the adult population, driving behaviour is an inherently complex process, with driving decisions affected by various factors, including network topography (e.g. type of road, number of lanes, curvature, gradient, etc.), traffic conditions (e.g. average speed, density, etc.), surrounding conditions (e.g. position and speed of the adjacent vehicles), path-plan of the driver (e.g. location of the next exit), features of the vehicle (e.g. acceleration and deceleration capabilities) and characteristics of the driver (e.g. age, experience, aggressiveness, impatience, stress level, and numerous others). Existing driving behaviour models address many of these factors, either fully or partially, where the effects of surrounding traffic conditions on the decisions of the driver in particular have received considerable attention from researchers. However, in most cases, the models do not adequately capture the sophistication of driver behaviour and the causal mechanism behind their observed decisions. Research in other realms, in the context of accident analysis and safety research in particular, has confirmed that driving behaviour is significantly affected by drivers' characteristics such as age, gender, education level and experience; individual traits such as aggressiveness, emotional instability, and venturesome/thrill-seeking attitude; as well as drivers' mental state/mood such as anger, tension, depression, cognitive workload and distraction.

The existing driving behaviour models being used in the leading microscopic simulators tend to overlook the effect of these driver specific factors on the decision framework and ignore the underlying heterogeneity in decision making of different drivers as well as the same driver in different contexts. The behavioural predictions from driving behaviour models that ignore the effects of driver characteristics and underlying heterogeneity are bound to contain significant noise as a result of the models' structural inability to uncover underlying causal mechanisms. Implementation of models failing to capture the full diversity of driving behaviour in traffic micro-simulation tools can lead to unrealistic traffic flow characteristics and incorrect representation of congestion.

The objective of this research will be to develop dynamic driving behaviour models that explicitly account for the effects of driver characteristics in his/her decisions alongside the effects of path-plan, network topography and traffic conditions. In a novel approach, the models will be calibrated by combining experimental data collected from the University of Leeds Driving Simulator (UoLDS) and actual traffic data collected using video recordings.

The topic warrants strong quantitative skills.


Search site