The invention provides a missing traffic state data filling method based on a generative adversarial network, and the method comprises the steps: firstly constructing an adjacency list of a detection road network, carrying out the random walk with bias of nodes in the adjacency list through employing a node2vec graph embedding algorithm, and obtaining an adjacent road traffic state matrix; and performing data normalization on the adjacent road traffic state data matrix by adopting a maximum and minimum standardization algorithm, constructing a generative adversarial network model, and determining a model structure of a generator and a discriminator. Defining loss functions of a generator and a discriminator in the generative adversarial network, selecting adjacent road traffic state data by using a sliding window method, taking the data as input of a generative adversarial network model, and minimizing the difference between the generated data and real data distribution through adversarial training, so as to obtain an adversarial network model; and generating the road traffic state of the missing road section. According to the method, the characteristics of traffic data are well combined, the data filling efficiency is improved, and compared with a traditional method, the method has better generalization performance.
本发明提供一种基于生成对抗网络的缺失交通状态数据填补方法,首先构建检测路网的邻接表,使用node2vec图嵌入算法对邻接表中的节点进行带有偏重的随机游走,得到邻接道路交通状态矩阵。对邻接道路交通状态数据矩阵采用最大最小标准化算法进行数据归一化,构建生成对抗网络模型,并确定生成器与判别器的模型结构。对生成对抗网络中的生成器与判别器的损失函数进行定义,使用滑动窗口法,对邻接道路交通状态数据进行选择,将其作为生成对抗网络模型的输入,通过对抗训练最小化生成数据与真实数据分布之间的差异,生成缺失路段的道路交通状态。本发明方法很好地结合了交通数据的特点,提高了数据填补的效率,相较于传统方法有更好的泛化性能。
Missing traffic state data filling method based on generative adversarial network
一种基于生成对抗网络的缺失交通状态数据填补方法
2023-12-29
Patent
Electronic Resource
Chinese
IPC: | G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
ST-FVGAN: filling series traffic missing values with generative adversarial network
Taylor & Francis Verlag | 2022
|Road traffic network state prediction based on a generative adversarial network
Wiley | 2020
|Traffic accident data quality enhancement method based on generative adversarial network
European Patent Office | 2023
|Road traffic network state prediction based on a generative adversarial network
IET | 2020
|