摘要
传统方法对无人水下航行器进行导航时,经纬度误差与实际误差相差较大。为了解决该问题,提出模式识别导航方法。统计满足特征组分辨尺度的各个特征,并计算离散度,设计特征选择流程,在对完成已经存在的特征进行更新后,对状态变量进行维护处理,对新加入的特征值进行坐标分析,并获取扩充后的状态向量,保证导航不会受到区域大小限制,设计具体航迹线确定方案,由此完成无人水下航行器智能导航。仿真实验结果表明,该方法比传统方法导航结果误差要小,且与实际误差一致,具有广阔应用前景。
The longitude and latitude errors are quite different from the actual errors when traditional methods are used to navigate unmanned underwater vehicles. To solve this problem, a pattern recognition navigation method is proposed. Statistical analysis satisfies the characteristics of feature group resolution scale, calculates dispersion, designs feature selection process, maintains state variables after updating existing features, carries out coordinate analysis of newly added feature values, and obtains expanded state vectors to ensure that navigation is not restricted by Region size, and designs specific route determination scheme. Thus, the intelligent navigation of unmanned underwater vehicle is completed. The simulation results show that the navigation error of this method is smaller than that of the traditional method, and it is consistent with the actual error, so it has broad application prospects.
作者
王莉
刘娜
WANG Li;LIU Na(School of computer engineering,Shangqiu University,Shangqiu 476000,China)
出处
《舰船科学技术》
北大核心
2019年第6期61-63,共3页
Ship Science and Technology
关键词
模式识别
无人水下航行器
智能导航
特征分量
状态变量
pattern recognition
unmanned underwater vehicle
intelligent navigation
feature components
state variables
