Abstract: To solve the dead reckoning navigation algorithm error accumulated problems, the underwater transponder auxiliary navigation calculates integrated navigation algorithm. Uses the current statistical model for traffic prediction model, a more accurate description of the vehicle movement state has been obtained. Using underwater transponder combined with navigation calculation algorithm, to air a reckon navigation error correction, in order to obtain a better filtering effect, the square root cubature Kalman filter has been introduced as the filtering algorithm of integrated navigation system, and compared with the EKF filtering algorithm. Simulation experiments show that the square root cubature Kalman filter has better filtering accuracy than the EKF filtering algorithm, and UTP/DR integrated navigation algorithm effectively avoids the navigation positioning error caused by the navigation error accumulated divergence problem, obtains the good effect of navigation and positioning.

Key words: Unmanned Underwater Vehicle（UUV）, navigation, Dead Reckoning（DR）, Underwater Transponder Positioning（UTP）, cubature Kalman

摘要： 针对UUV水下作业时航位推算存在导航误差积累的问题，研究了水下应答器辅助航位推算组合导航算法。采用“当前”统计模型作为航位推算模型，更准确地描述了航行器的运动状态。利用水下应答器与航位推算算法相组合，对航位推算导航误差进行校正。为获得更好的滤波效果，采用平方根容积卡尔曼滤波算法作为组合导航系统的滤波算法，并将其与EKF滤波算法进行比较。仿真实验表明，平方根容积卡尔曼滤波算法较EKF算法具有更好的滤波精度；UTP/DR组合导航算法有效避免了因导航误差积累而导致的导航定位误差发散问题，获得了较好的导航定位效果。

关键词: 无人水下航行器（UUV）, 导航, 航位推算, 水下应答器定位, 容积卡尔曼