基于角增量和矢量模值观测的光纤陀螺温度漂移参数分段估计

FOG Offset Thermal Drift Coefficient Piecewise Estimation Based on Angular Increments and Vector Norm Value Observation

光纤陀螺捷联惯导系统被广泛应用于航空、航天、航海及陆地车辆定位定向等领域,对光纤陀螺输出误差进行补偿是提高导航精度的有效手段。温度漂移和常值零偏是影响光纤陀螺精度的两个主要误差来源,对角增量输出式三轴光纤陀螺捷联惯导系统的陀螺温度漂移及常值零偏误差参数估计方法进行了研究。针对光纤陀螺的温度漂移,提出了一种基于角增量的分段最小二乘估计方法,根据不同温度区间的特征使用低阶模型即可进行误差建模,估计结果相比整体估计方法更加精确,同时推导了各个温度段参数的边界条件,保证了温度漂移模型在不同温变速率条件下的连续性。针对三轴陀螺输出中包含的常值零偏,提出了一种基于地球自转角速度矢量模值观测的方法,可在不依赖高精度转台等外部基准设备的条件下对光纤陀螺零偏进行估计,可适用于高纬度地区及极区环境下的外场标定。通过温箱静置升温实验,对光纤陀螺惯导系统三轴角增量陀螺进行了温度漂移和零偏的估计与补偿,验证了提出方法的有效性。

Fiber optic gyroscope(FOG)strapdown inertial navigation system(SINS)is widely used in aviation,aerospace,voyage and land vehicle positioning.Compensation of FOG output error is an effective means to improve navigation accuracy.Thermal drift and constant bias are the two main error sources affecting FOG accuracy.Therefore,thermal drift and constant bias error estimation method are developed for three-axis FOG with angular increment output in SINS.A piecewise low order thermal drift error model using least squares estimation is proposed.Considering thermal drift characteristics of different temperature segments,the new model is more accurate than the traditional one which regards the temperature interval as a whole segment.To guarantee the continuity of the thermal drift error model between adjacent temperature segments under different temperature changing rates,the boundary conditions of each temperature segment are derived.A vector norm value observation method is presented to estimate the constant bias error in the three-axis gyroscope taking advantage of the earth’s rotation angular velocity vector and gravity vector as references,while the external reference device such as a high-precision turntable is not needed.Beside this outfield calibration method can be applied in high latitude areas even in polar regions.The effectiveness of the proposed methodology for FOG offset thermal drift coefficient piecewise estimation has been validated through a stationary heating test.