Abstract

The focus of this study is the interaction between cars at multi-lane roundabouts. On such roundabouts it is important that cars change lanes at the appropriate time to prevent accidents occurring. For example, an accident will occur if either the car in the outside lane attempts to leave the roundabout early or if the car on the inside lane stays on the roundabout longer than expected. Both of this actions could cause the car in the outside lane to go into the car on the inside lane but for different reasons. This study will aid the investigation into this interaction between the cars and provide an insight into how various actions can be taken by a car to either prevent an accident or cause one.

This paper will cover two principal areas, these are concurrent system modelling and educational technology. The modelling of the cars movements creates the concurrent aspect of the study. The cars could be thought of agents in the system in which they all act independently of each other and yet need to be aware of each other’s actions such that they can all achieve their goal successfully (leaving the roundabout at the correct exit while avoiding crashing). The educational aspect is derived from a user’s interaction with the model and their subsequent learning experience. A user will be able to set a number of parameters within the model allowing them to carry out a thorough investigation of how different factors are linked, i.e. the dependencies within the model. The type of user who could use this model to learn could be anyone, however, it is intended for either people learning to drive who wish to gain a further understanding of the importance of their actions on the road or an accident investigator wishing to analyse a particular situation to find possible causes of accidents.