火箭弹简易捷联惯导系统的姿态解算修正方法

Attitude Algorithm for Simple Strap - down Inertial Navigation System in Rocket Ammunition

在导航运动系统精度问题的研究中,火箭弹上简易捷联惯导系统的速率陀螺测量误差和漂移较大,如果直接对速率陀螺积分解算姿态可能导致较大偏差。而传统的滤波方法将弹体的姿态运动认为是近似线性的变化,对高频运动的火箭弹建模会产生较大的模型误差。为有效提高滤波模型的精确性和组件测量精度,提出了一种根据当前统计模型的姿态运动模型滤波方法,通过对运动状态的有限转移建模可以较好地描述火箭弹姿态的高频运动方式,并进行仿真。仿真实验证明,统计模型建立的滤波方法对火箭弹简易捷联惯导系统的姿态解算结果要优于传统的基于线性拟合模型的滤波方法。

The measurement error and drift of simple strap-down inertial measurement unit in the rocket ammunition are serious,and there may be a large attitude deviation if the output of velocity gyroscope is integrated directly.However,the traditional filtering methods view the attitude motion of the body as a near-linear change,which can increase the model error when building the high-frequency motion model of rocket ammunition.To improve the precision of filtering model effectively,this paper presented a new filtering method for attitude motion based on the current statistical model,and with this method,the high-frequency motion of rocket ammunition can be well described through modeling a motion state with finite transformation.Finally,the simulation results show that,this new method is superior in the attitude solution result for simple strap-down inertial measurement unit in the rocket ammunition to the traditional filtering method based on linear fitted model.