论文部分内容阅读
在传递对准精度评估中,由于惯导系统模型参数与实际的物理过程存在偏差、系统量测噪声的特性不确定等因素,采用常规的卡尔曼滤波算法往往不能获得理想的滤波效果。为满足传递对准滤波估计的精度和稳定性要求,设计了一种新型平方根自适应滤波算法:即将简化的时变噪声统计估计器串联于平方根滤波过程的时间更新和量测更新之间,使平方根滤波与自适应滤波方法相结合,从而实现抑制滤波过程中的模型发散和计算发散的目的。仿真结果表明,该算法与传统Kalman滤波算法、简化Sage-Husa自适应算法相比,不仅增强了滤波的收敛能力,而且大大提高了估计精度,能够有效地适用于传递对准精度评估。
In the evaluation of transmission alignment accuracy, the conventional Kalman filtering algorithm often can not obtain the ideal filtering effect because of the deviation of the parameters of the inertial navigation system from the actual physical process and the uncertainty of the system measurement noise characteristics. In order to meet the requirement of accuracy and stability of pass-by-pass filter estimation, a new type of square-root adaptive filtering algorithm is designed: a simplified time-varying noise statistical estimator is connected in series between the time update and the measurement update of the square root filter process, Square root filter and adaptive filtering method combined to achieve the suppression filter in the process of model divergence and the purpose of calculating divergence. Simulation results show that compared with the traditional Kalman filter and Sage-Husa adaptive algorithm, the proposed algorithm not only enhances the convergence ability of the filter, but also greatly improves the estimation accuracy. It can be effectively applied to the accuracy of transfer alignment.