基于SUKF算法的组合导航方法

合成孔径声纳（SAS）运动补偿是其信号处理的重要环节，运动补偿需要水下导航数据，文中研究了基于捷联惯导（SINS）和多普勒声速剖面仪（ADCP）等设备的水下导航方法。针对大姿态误差角情况下的导航，论文建立了基于四元数的SINS／ADCP的非线性误差模型，并采用比例无迹卡尔曼滤波（SUKF）滤波算法进行了仿真。仿真结果表明，在SAS复杂运动情况下，采用非线性误差模型的SUKF方法优于传统线性误差模型的卡尔曼滤波方法，在一定程度上提高了SAS的导航精度。

Integrated Navigation Method Based on SUKF Algorithm

Synthetic aperture sonar（SAS） imaging is based on the straight track model with constant velocity. The motion error estimation for ascertaining the ideal track is usually achieved by the integrated navigation with high precision. For the larger attitude error angle, nonlinear error models of strapdown inertial navigation system/acoustic Doppler current profiler（SINS/ADCP） based on quaternion error are established. The simulations are performed with the scaled unscented Kalman filter （SUKF） algorithm used in the nonlinear error models and the Kalman filter （KF） algorithm in the linear error models,respectively. The results indicate that the SUKF algorithm in the nonlinear models is superior to the KF algorithm in the linear error models in the respect of navigation accuracy.