在惯组系统级标定过程中, 由于系统模型尤其是噪声统计特性的不确定性, 常常造成较大的估计误差, 严重时甚至导致滤波器发散。针对此问题, 采用Two-Stage滤波思想, 研究随机噪声干扰下系统不确定性偏差的最优滤波器设计(OTSKF), 并在此基础上提出一种基于最优TSKF算法的快速次优滤波算法。理论分析表明该算法具有较小的运算量、良好的收敛性及抗扰动性。最后, 将该算法应用于惯组系统级标定, 通过一组自动化标定方案, 实现了惯组的在线标定, 实验结果验证了该算法的有效性。

In systematic calibration of IMU,the inaccurate system model and inexact stochastic information may degrade the performance of the Kalman filter,or even cause the filter to diverge.To solve such problems,the design and implementation of optimal filters based on Two-Stage Kalman filter for systems with unknown bias was discussed.A suboptimal algorithm was proposed for linear stochastic systems with a random bias,and theoretical analysis showed that the algorithm has less computation cost,fine convergence performance and good robustness.To verify the performance of it,the proposed algorithm was applied to the online calibration of IMU though a new automatic systematic calibration scheme.The result proved the effectiveness of it.