基于移动Kriging插值算法的水下地形构建技术研究

An underwater terrain construction method based on moving-Kriging interpolation algorithm

针对Kriging插值算法在地形构建中对真实观测数据进行光滑处理时可能产生较大误差的问题,在简述其基本原理的基础上,提出了移动Kriging插值算法,并详细阐述了该算法的原理和步骤。运用该算法对类Sphere、Rastrigin函数分别模拟的平坦和起伏地形进行插值重构仿真测试,并对武汉木兰湖湖底真实地形数据进行了验证比较,结果表明:该算法在水下地形构建中能够获得较好的插值效果,很好地保持了地形特征,且能获得更小残差的水下地形数据,在水下数字地图制作中具有较好的应用价值。

In order to study the problem of the large errors caused by Kriging interpolation algorithm for smoothing the real observation data, a moving-Kriging interpolation algorithm is proposed based on the basic principle of Kriging interpolation algorithm, and its implementation process and application is introduced in detail. The algorithm is used to interpolate and reconstruct the flat and undulating terrain simulated by the similar Sphere and Rastrigin functions respectively, and the real terrain data of the Mulan Lake in Wuhan is verified and compared. The result indicates that better interpolation effect in underwater terrain construction can be achieved using the algorithm while the topographical features are well maintained. Moreover, the low-residual underwater terrain data obtained have good prospect in underwater digital mapping.