The invention relates to the technical field of intelligent ship driving, in particular to a maritime multi-ship encounter graph structured learning method based on a graph convolutional neural network, which comprises the following steps of: 1, screening and identifying two-ship encounter data from AI (Artificial Intelligence System) big data; 2, obtaining multi-ship meeting data based on cross matching of the meeting data of the two ships; step 3, performing graph structuring on multi-ship encounter data; and step 4, constructing a graph convolutional neural network. The method has great contribution to marine operation efficiency and safety under complex traffic conditions, and can be potentially applied to autonomous navigation of an unmanned ship cluster under a future marine man-machine mixed traveling condition, so that the collision intention of a ship driver can be better known under the scene that multiple ships meet.
本发明涉及船舶智能驾驶技术领域,特别是一种基于图卷积神经网络的海上多船会遇图结构化学习方法,包括以下步骤:步骤1、从AI S大数据中筛选识别两船会遇数据;步骤2、基于两船会遇数据交叉匹配得到多船会遇数据;步骤3、对多船会遇数据进行图结构化;步骤4、构建图卷积神经网络。该发明对在复杂的交通状况下的海上运营效率和安全性有很大贡献,并且可以潜在地应用于未来海上人机混行条件下的无人船集群自主航行,在多船会遇的场景下,以更好地了解船舶驾驶员的碰撞意图。
Marine multi-ship encounter graph structured learning method based on graph convolutional neural network
基于图卷积神经网络的海上多船会遇图结构化学习方法
2024-03-22
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Multi-scale traffic flow prediction method based on graph convolutional neural network
European Patent Office | 2021
|Multi-graph hierarchical road network characterization method based on graph neural network
European Patent Office | 2023
|European Patent Office | 2025
|