基于测量分量加权融合的GPS／SINS船舶组合导航算法

GPS/SINS Ship Integrated Navigation Algorithm Based on Measurement Component Weighted Fusion

具有执行简便和高精度优点的测量值融合技术已经在船舶组合导航中得到了广泛应用。但由于船舶组合导航系统中各子系统通常提供的是不等同维的测量数据,加之传统测量值加权融合存在各传感器测量等同维数的约束,使得该类算法在船舶组合导航中的应用能力和使用范围受到了很大的限制。针对上述问题,以GPS和SINS组成的、具有不同维数测量的船舶组合导航系统为研究对象,提出了一种新的基于测量分量加权融合的船舶组合导航算法。该算法的基本思想是寻找一个变换矩阵将具有不等同维数的测量扩维成满状态维数的矩阵,然后考虑扩维后的噪声方差是奇异矩阵,从而采用测量分量对应加权进行融合。与传统的测量加权融合方法相比,该方法不仅具有更广的应用范围,而且保持了与传统的集中式扩维融合相同的高精度性能。算法的性能分析和计算机仿真证明了新算法的有效性和优越性。

Measurement fusion technology has been widely applied to ship integrated navigation system. Traditional measurement weighted fusion （TMWF） has its limitation for it requires equal dimensional measurement, while different subsystems in the integrated navigation system provide different dimensional measurements. Take the GPS/SINS ship integrated navigation system as a typical example, a new integrated navigation algorithm based on generalized measurement weighted fusion （GMWF） is presented. The idea of the new algorithm is that a transform matrix is built to uniform different dimensional measurements （measured by GPS） so that fusion can be realized by the measurement component weighted method. Proved by the computer simulation and performance analysis, the integrated navigation method based on GMWF has a high fusion precision and a wider application.