一种多地球物理特征匹配自主导航方法

An Approach to Autonomous Navigation Using Matching of Geophysical Features

针对地球物理场具有多个特征和匹配可操作性较强的特点,文中提出了一种多地球物理特征匹配的自主导航方法,研究了导航方法的实现方式和对应的匹配算法。导航方法通过同步测量航迹上的地磁场的多个特征和重力梯度,采用固定点数的滑动窗口构造测量序列后与基准图匹配,最后对匹配结果做融合,得到实时位置信息。匹配算法设计了基于"初始位置+初速度+加速度"实值编码的遗传算法作为搜索策略,采用平均平方差准则(MSD)作为每个特征的匹配相似性度量,并使用加权最大值原则融合所有特征的相似性度量。最后由仿真算例验证了方法的可行性,表明了导航方法具有良好的匹配精度,可用于低空、低速运动载体的导航。

Section 1 of the full paper discusses the feasibility of the autonomous navigation using matching of geophysical features. Section 2 explains the principles of the autonomous navigation using matching of geophysical features; eq. （4） is proposed by us; its core consists of： （1） each feature of geomagnetic field and the gravity gradient in the flight path of a flight vehicle are measured synchronously; （2） each measurement sequence is constructed and matched against the corresponding reference map; （3） the matching results of all features are fused and then the location information is obtained. Section 3 explains the design of the genetic algorithm we use; its core is that the genetic algorithm based on the real acceleration value code is designed as the search algorithm of the matching algorithm; the matching algorithm applies the mean square difference （MSD） algorithm as the similarity criterion for each feature and uses the weighted maximum value principle to fuse the similarity values of all features. The simulation results, presented in Figs. 2, 3 and 4, show preliminarily that the positioning error obtained with this approach is small ; hence it can be applied as an approach to the autonomous navigation of the low-altitude and low-speed vehicles.