基于投影寻踪的多维地磁匹配融合算法

单一地磁特征匹配算法容易受到特征量区域缓变特性的影响,从而降低了算法的定位精度。针对单一特征量匹配算法的区域性不可靠问题,提出了一种多维地磁匹配融合算法;该算法在建立多维地磁匹配特征量投影寻踪分类模型的基础上,采用组织进化数值优化算法对模型进行求解,算法根据采样数据的特点自适应获得各匹配特征量的定位贡献权重(可信度),将经过加权融合的匹配位置作为最终定位位置,避免了单特征量匹配的区域性不可靠问题。仿真结果表明,采用多维地磁匹配融合算法较单一特征匹配算法大大减小了全局匹配定位误差,验证了算法的有效性。

Multidimensional Geomagnetic Matching Fusion Algorithm Based on Projection Pursuit

The algorithm of geomagnetic matching of single characteristic parameter is influenced by the slow changing of characteristic parameter easily,and reduces the orientation precision of the matching.A multidimensional geomagnetic matching fusion algorithm was presented,in order to solve the territorial uncertainty problem of the single parameter matching method;the multidimensional geomagnetic matching fusion was performed by the algorithm presented above,on the basis of the characteristic parameter's project...