An Arch-Bridge Topology-Based Expressway Network Structure and Automatic Generation

Abstract

The rapid generation and automatic updating of road network topology information have immense practical significance and application value for intelligent expressways. Therefore, this work proposes a novel arch-bridge topology based on an electronic toll collection (ETC) system to redefine the structure of the expressway network. On this basis, discrete ETC data generate corresponding trajectories according to the arch-bridge topology structure, and an initial topology candidate set is generated. Furthermore, the characteristics of abnormal topologies, such as loop, reverse, missing, and opposite topologies, are deeply explored and analyzed, and targeted constraint conditions are designed to optimize the initial topology candidate set. Finally, experiments and evaluations are conducted on transaction data collected from an ETC system in a certain province of China. The experimental results show that the proposed method is practical and has significant results in terms of evaluation metrics and efficiency. Specifically, the recall, precision, and F1 are 0.982, 0.966, and 0.974, respectively. Additionally, it takes less than 2 ms on average to generate one topology. The proposed method can efficiently and accurately generate the topology information of an entire expressway network.