Vehicle navigation system (VNS) is a world acknowledged efficient way of solving the urban traffic problem, and road network database is one of the core component parts in the system. In order to improve performance of VNS, this paper conducts a research on data organization technique of road network, and proposes a novel data model of road network, which includes spatial index and topology structure. Firstly, some common spatial index structures, such as regular grid index, R-tree index and quad-tree index, are studied, and a novel spatial index structure based on “hierarchical quad-tree and R-tree” index of two ranks is proposed. This structure can greatly reduce the access time of map data and raise index performance. Secondly, a topology model of road network based on arc-relation is presented, which can effectively solve turn penalty problems in the conventional topology model based on vertex-relation and represent real road network. Thirdly, a real topology storage structure using adjacency list is designed and a corresponding creating algorithm is put forward. Finally, the tests on the practical system prove that the proposed model effectively support kinds of data-processing and greatly raise the performance of the system.
Intelligent Vehicle Navigation System Based on Visual Detection and Positioning
