基于参数识别的SINS/DVL初始对准方法

Initial alignment method for SINS/DVL based on parameter identification

针对多普勒测速仪(DVL)辅助捷联惯导系统行进间对准时易受DVL量测噪声的影响,提出一种基于参数识别的SINS/DVL初始对准方法。首先,建立了基于DVL辅助的SINS行进间初始对准观测矢量模型,分析了DVL量测噪声对观测矢量的影响;然后,研究了观测矢量变化规律,建立了观测矢量参数识别模型,利用建立的参数识别模型,设计了基于自适应卡尔曼滤波的参数识别算法,并对观测矢量进行了重构,减小了DVL量测噪声对观测矢量的影响;最后,设计了仿真与跑车实验。实验结果表明,所提出的参数识别算法可以有效抑制DVL量测噪声对初始对准结果的影响。相较于传统方法,在载体运动条件下实现对准误差标准差小于0.1°。

To suppress the measurement noises of DVL,a parameter identification method is proposed for SINS/DVL initial alignment.Firstly,an observation vector model for in-motion alignment with DVL aided is constructed,and the influence of DVL measurement noise on observation vector is analyzed.Secondly,the parameter model of the observation vector is studied and constructed.Based on the constructed model of observation vector,a parameter identification method with an adaptive Kalman filter is designed for estimating the optimal parameters,which are used to reconstruct the new noises-free observation vector. Finally,a simulated and field test is designed for verifying the proposed method.The simulation results demonstrate that the measurement noises are eliminated effectively,and the standard deviation of the alignment errors is smaller than 0.1° when the vehicle is in-motion.