基于Unscented卡尔曼滤波算法在海底地形辅助导航中的应用

Application of an unscented Kalman filter algorithm in the seabed terrain aided navigation

为了解决SITAN海底地形辅助导航算法将非线性过程线性化,容易产生不稳定甚至滤波发散的问题,提出了一种基于Unscented卡尔曼滤波的海底地形辅助导航算法,该算法无需对非线性系统的状态方程和观测方程进行线性化,直接采用UT变换(unscented transform,UT)近似非线性函数的概率密度分布,并求取变换后的点集的均值和方差。利用海底地形SEAMAPnew Hill100为真实海底地形数据对算法进行了测试。实验结果表明,该算法较传统算法有较强的精确性和鲁棒性。

In order to solve the problem that the linearization of the nonlinear process of the SITAN seabed terrain aided navigation algorithm is prone to instability and even the filtering divergence,a seabed terrain aided navigation algorithm based on unscented Kalman filter is proposed. This algorithm does not need nonlinear state and observation equations to be linearized,directly using the unscented transform（ UT） to approximate probability density distribution of nonlinear functions,and obtains the mean and variance of the transformed point set. It uses the seabed terrain SEAMAPnew Hill100 as real terrain data to test this algorithm. The experimental results show that the algorithm is more precise and robust than the traditional algorithms.