测绘学报

• 学术论文 •    

水下重力异常相关极值匹配算法研究

李姗姗1,3,吴晓平2,马彪1,3   

  1. 1. 信息工程大学测绘学院
    2. 解放军信息工程大学测绘学院一系
    3. 信息工程大学测绘学院
  • 收稿日期:2010-07-13 修回日期:2011-02-27 出版日期:2011-08-25 发布日期:2011-08-25
  • 通讯作者: 李姗姗

Research on the Correlative Extremum Matching Algorithm Using Underwater Gravity Anomalies

  • Received:2010-07-13 Revised:2011-02-27 Online:2011-08-25 Published:2011-08-25

摘要: 重力辅助惯性导航是利用地球物理特征信息—重力来修正水下潜器惯性导航误差,匹配算法是其关键技术。本文基于平均平方差最小准则构造了差分降权相关目标函数模型,提出了由于受干扰误差影响导致源于正确位置出现多个有效位置的概率数据关联滤波重力匹配算法,与选择最近的唯一位置相比更接近于实际的真实位置,提高了算法的可靠性与稳健性;分析探讨了序列采样间隔对匹配精度的影响。通过实验区仿真匹配,结果表明,当选择适当的采样间隔与采样长度,本文提出的算法能有效修正导航误差。

Abstract: Gravity-aided inertial navigation is a technique that makes use of gravity to correct inertial navigation error of underwater vehicles. Matching algorithm is vital in the process of gravity-aided inertial navigation. An object function model of differential descending weight-correlation was put forward based on the principle of mean square deviation. Besides, a probabilistic data association filter algorithm was proposed for multi-available positions that resulted from the disturbing errors of the real position. Compared with choosing the unique position which is closest to the actual one, PDAF promotes the reliability and the robustness of the algorithm. Then, the effect of the serial sampling intervals on the matching precision was discussed. Through simulations in the experimental area, the results showed that navigation errors were corrected effectively with proper sampling intervals and sampling lengths based on the algorithm of this paper.