强机动条件下的弹载深组合ARCKF滤波方法研究

Research on the Method of Missile-loaded Deeply-Integrated ARCKF Filtering under Strong Maneuver

针对弹道导弹飞行中GNSS/SINS深组合导航系统呈现的强机动、非线性特性,引入了基于三阶球面-径向准则的容积卡尔曼(CKF)非线性滤波方法。同时,提出了一种自适应抗差容积卡尔曼滤波(ARCKF)算法,该算法运用抗差M思想,调节量测噪声阵,以抵御系统观测异常扰动,采用自适应因子对协方差阵进行调节,进一步处理动态扰动引入的误差。实验结果表明,该滤波算法有效提高了组合导航的动态性能,在加速度达到60g的强机动仿真环境中,仍能保持较高的导航精度和跟踪性能。

Aiming at the strong maneuvering and nonlinear characteristics of GNSS/SINS deeply integrated navigation system in ballistic missile flight, the cubature Kalman filter （CKF） based on third-degree spheri- cal-radial cubature rule is introduced in this paper. Meanwhile, an adaptive robust cubature Kalman filter （ARCKF） is proposed. The robustness M is used in this algorithm to adjust the measurement noise matrix to resist system disturbance, and the adaptive factor is used to adjust the covariance matrix to further deal with the error introduced by the dynamic disturbance. The experimental results show that the dynamic perform- ance of integrated navigation system can be effectively improved by using proposed algorithm, and high nav- igation precision and tracking performance can be maintained under the strong maneuver simulation envi- ronment with acceleration up to 60g.