TURAS: A Personalised Route Planning System

Abstract

Route planning has always provided a rich vein of research for the artificial intelligence community. However, to date the majority of this research has focused on the generation of optimal routes using shortest path algorithms to minimise distance traveled. The problem of planning high quality realistic routes is difficult for several reasons. Firstly, real maps rarely contain the sort of information that is useful for constructing realistic, high quality routes (real-time traffic information or road quality data). Secondly, the notion of “route quality” is notoriously difficult to define and is likely to change from person to person. Consequently, in real-world route planning situations, the shortest route is rarely the best route for a given user.