【摘要】 为了提高捷联惯性导航系统在线标定的精度和实时性,根据模型预测滤波算法和Sage_Husa自适应卡尔曼滤波算法的优点,提出了一种新的自适应模型预测组合滤波算法。该算法首先利用模型预测滤波算法估计出系统模型误差,并对系统状态方程实时修正,以减小系统模型误差对导航精度的影响;然后利用简化的自适应滤波算法对量测噪声在线调整,修正噪声统计特性,以提高滤波精度。将提出的算法进行在线标定仿真实验,并与传统的卡尔曼滤波在线标定算法进行比较,结果表明,提出的自适应模型预测组合滤波算法能有效完成在线标定,且标定精度和收敛速度均优于传统方法。更多还原

【Abstract】 In order to improve the accuracy of online calibration and improve the real-time performance of the strapdown inertial navigation system(SINS), a new filtering algorithm is proposed, which combines model predictive filtering(MPF) with Sage-Husa adaptive Kalman filtering(AKF) to take use of their advantages. Firstly, the algorithm takes the advantages of MPF to estimate the model error and update the system state equation to inhibit the influence of the model error on the navigation accuracy. Secondly, the simplified AKF is used to adjust the measurement noise and modify the noise statistic characteristics in order to improve the filtering accuracy of the system. The proposed algorithm is applied to online calibration and compared with AKF. The simulation results demonstrate that the online calibration can be accomplished efficiently, and the accuracy is better than those of traditional approaches.更多还原