基于UKF的捷联惯导大失准角初始对准方法

摘要

The feature of propagate of large misalignment of strapdown inertial navigation system is nonlinear. In order to improve initial alignment algorithm of SINS with large misaligmnent, this article proposes a method based on Unscented Kalman Filter. Under the assumptions of the additive noises in both state process and measurement process and the linear measurement mode, the simplified Unscented Kalman Filter (UKF) is presented for the SINS initial alignment model of AQE with large misalignment. The method of multiple feedbacks to SINS with the UKF state estimates is proposed to increase the alignment accuracy in the process of initial alignment. The simulation results demonstrate that the SINS initial alignment can be accomplish aided by GPS, even if azimuth of error reaches 20°, and using three precision alignment, the high alignment accuracy is available.%研究自主水下航行器,位置速度匹配中的误差需要补偿.捷联惯导大失准角误差传播特性是非线性的.为了提高捷联惯导大失准角初始对准精度,提出了一种基于UKF的捷联惯导初始对准的方法.由于对大失准角下SINS的非线性误差模型和系统噪声和观测噪声为加性的特点,给出了一种简化的Unscented卡尔曼滤波递推算法,并提出在精对准阶段采用多次精对准提高大初始失准角下SINS对准精度进行仿真.仿真结果表明,在GPS辅助下,使捷联惯导方位角误差减少,采用改进方法,也可获得较高的对准精度.