摘要
针对狭水道航行等对位置信息精度与稳定性要求较高的场景,提出一种基于长短期记忆(Long Short Term Memory,LSTM)神经网络的舰船航行位置预测模型。完成模型构建的同时,结合导航系统、操控与推进系统、气象水文系统的航行试验历史数据,对模型进行了训练与测试,实现了对未来时刻舰船航行位置的预测。试验结果表明,舰位预测模型根据船舶航行历史数据对未来时刻航行位置的预测具有较高的准确性与稳定性,能够对船舶在狭水道航行等特殊场景下的安全航行提供辅助与支持。
A ship navigation position prediction model based on LSTM neural network is proposed for narrow channel navigation scenarios that require high accuracy and stability of position information.At the same time,the model is trained and tested based on the historical data of navigation system,control and propulsion system and hydro-meteorological system,and the prediction of ship's navigation position in the future is realized.The test results show that the ship position prediction model has high accuracy and stability in predicting the future navigation position according to the ship navigation history data,and can provide assistance and support for the navigation safety of ships in special scenarios such as narrow channels.
作者
韩旭
孙文本
周智楠
苑海静
HAN Xu;SUN Wenben;ZHOU Zhinan;YUAN Haijing(Tianjin Navigation Instruments Research Institute,Tianjin 300131)
出处
《舰船电子工程》
2023年第9期58-61,共4页
Ship Electronic Engineering
关键词
导航系统
LSTM
神经网络
位置预测
navigation system
LSTM
neural network
position prediction