The invention discloses a road network dynamic traffic distribution method based on an Encoder-Decoder deep neural network. The method comprises the steps of building an Encoder-Decoder model through employing an LSTM neural network model based on vehicle track data, inputting road network traffic multi-period traffic flow OD, outputting road section flow and intersection entrance lane steering flow, introducing an Attention mechanism in a model building process, and carrying out the dynamic traffic distribution, and obtaining a nonlinear mapping relationship between the road network traffic demand and the actual road flow. The result precision of the obtained distribution model is high, the defects that a traditional method is low in precision, poor in real-time performance and the like are overcome, and powerful support is provided for solving the key technical problem of traffic.
本发明公开了基于Encoder‑Decoder深度神经网络的路网动态交通分配方法,依托于车辆轨迹数据,采用LSTM神经网络模型构建Encoder‑Decoder模型,输入为路网交通多时段车流OD,输出为道路断面流量以及交叉口进口道转向流量,构建模型过程引入Attention机制,进行动态交通分配,得到路网交通需求与实际道路流量之间的非线性映射关系。本发明得到的分配模型结果精度较高,解决了传统方法存在的精度低、实时性差等弊端,为解决交通关键技术问题提供有力支撑。
Road network dynamic traffic distribution method based on Encoder-Decoder deep neural network
基于Encoder-Decoder深度神经网络的路网动态交通分配方法
2021-04-09
Patent
Elektronische Ressource
Chinesisch
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES |
UNSUPERVISED ENCODER-DECODER NEURAL NETWORK SECURITY EVENT DETECTION
Europäisches Patentamt | 2020
|Unsupervised encoder-decoder neural network security event detection
Europäisches Patentamt | 2022
|UNSUPERVISED ENCODER-DECODER NEURAL NETWORK SECURITY EVENT DETECTION
Europäisches Patentamt | 2023
|Road network dynamic traffic distribution method based on traffic flow prediction
Europäisches Patentamt | 2024
|