| 基于Elman网络的舰船组合导航系统误差补偿研究 |
| |
| 引用本文： | 马野,王孝通,李博.基于Elman网络的舰船组合导航系统误差补偿研究[J].中国航海,2004(4):54-56,63. |
| |
| 作者姓名： | 马野 王孝通 李博 |
| |
| 作者单位： | 海军大连舰艇学院,辽宁,大连,116018 |
| |
| 摘 要： | 为了提高卡尔曼滤波估计精度，提出了一种基于回归神经网络补偿卡尔曼滤波器估计误差的方法。根据Elman网络与非线性ARMA模型工作原理的相似性，利用Elman网络做误差估计器，补偿卡尔曼滤波器的估计精度。实际舰船航行数据仿真测试表明，该方法有效可行。
|
| 关 键 词： | Elman网络 回归神经网络 行数据 误差补偿 估计精度 仿真测试 估计误差 组合导航系统 卡尔曼滤波器 ARMA模型 |
| 文章编号： | 1000-4653(2004)04-0054-03 |
Research on Ship Integrated Navigation System Compensation Errors Based on Elman Networks |
| |
| MA Ye,WANG Xiao-Tong,LI Bo.Research on Ship Integrated Navigation System Compensation Errors Based on Elman Networks[J].Navigation of China,2004(4):54-56,63. |
| |
| Authors: | MA Ye WANG Xiao-Tong LI Bo |
| |
| Abstract: | A method based on recurrent neural network compensation Kalman's evaluation error is proposed in order to enhance the evaluation precision. According to the similarity of working principles of Elman network and non-linear ARMA model, Elman network is utilized to be an error estimator to compensate the estimation precision of Kalman filter. The simulation tests using the data of real ship with MATLAB indicate that this method is effective and feasible. |
| |
| Keywords: | Ship Navy vessel engineering Recurrent networks Research ARMA model Error Neural networks Integrated navigation Computer simulation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
