旋转式惯导系统光纤陀螺在线自标定方法

摘要

针对旋转式惯导的光纤陀螺标定问题，基于旋转式惯性测量单元设计了标定光纤陀螺误差参数的最小二乘估计算法和相应的在线标定方案。基于标定结果可改进惯导系统的初始对准：对基于Kalman滤波的初始对准，可进行模型降阶，从而降低滤波运算量；对基于参数辨识的初始对准，根据标定的参数可以计算出陀螺的等效零偏，降低方位失准角初始值的近似误差。由于各陀螺输出信号处理过程相互独立而且具有对称性，在硬件实现过程中对递推最小二乘解算模块分时复用，可以节省2/3运算逻辑资源，其中的三角函数和除法器可进一步通过分时复用进行优化。本文提出的标定方法精度高，对初始对准精度有改善效果，同时易于硬件实现，因此工程实用性强。%To address the issue of the calibration of the fiber optical gyros of the rotational inertial navigation sys⁃tem, we designed the recursive least square (RLS) estimation algorithm based on rotational inertial measurement u⁃nit and related online calibration method to calibrate the error parameters of fiber optic gyros (FOG) online. The in⁃itial alignment of the inertial navigation system can be improved based on the calibration results. For the initial a⁃lignment based on Kalman filter, a reduced⁃order model was proposed to decrease the amount of filtering calcula⁃tions;For the initial alignment based on parameter recognition, the calibrated parameters can be used to calculate the equivalent gyro bias, and thereby, decrease the approximation error of initial azimuth error. Because the proce⁃dure processing signal outputted by each gyro is independent and is of symmetry, applying time division multiple⁃xing to the RLS solver in hardware implementation, the resources for RLS solver can be reduced by 2/3. And the internal divider and trigonometric function solver can be further optimized by time division multiplexing. The pro⁃posed calibration method is of high precision, helps to improve the initial alignment, and is apt to be implemented with hardware, and thus, is of good practicality in engineering application.