A data model for complex road network for vehicle navigation systems

Abstract

Vehicle Navigation Systems (VNS) is a core component of Intelligent Transport Systems (ITS). VNS can provide useful traffic information to drivers for facilitating the travel. In addition to display the current position of the vehicle, it has many other functions such as pre-trip planning, ad-hoc routing, and route guidance. Thus, a comprehensive navigable database of road network is very important to VNS. Road network database provides a foundation for all ITS applications, e.g. to support vehicle positioning enhancement, route guidance, and optimum path computation. Now, the difficult task is to map a complex road network into a single database. The difficulty rests in the fact that there are special traffic features in the road network to make the Hong Kong road network very complicated. There are a lot of special traffic features to guide and smooth the demanding traffic flow, such as legal divider which causes difficulties in modelling the network for vehicle navigation. Presently, three types of road models are in use, i.e. the Road Segment Model, Traffic Direction Model and Lane Model. All these three models are analysed and tested using the complex network of Hong Kong. The results show that none of them is capable of accommodating such complex network for effective route determination. As a result, an Adaptive Model is developed to provide a solution for modelling the complex road network for VNS. The adaptability is achieved by using a set of guidelines for storing the network connectivity and traffic restriction in a navigable database. This new model is experimented using the complex Hong Kong road network. The results show this model is capable of accommodating such very complex network. The resulting database is very efficient and effective for VNS, especially for route planning.