捷联惯导飞行器旋转矢量姿态优化仿真研究

Study and Simulation of Improved Rotation Vector Attitude Algorithm for Strapdown Inertial Navigation Aerocraft

在捷联惯导飞行器姿态优化的研究中,为了克服不可交换误差对计算精度的影响,目前常采用基于旋转矢量的姿态算法,利用陀螺仪的多次采样数据对不可交换误差进行补偿。因此采样的子样数越多,姿态计算精度越高。但是随着子样数目的增加,运算量及数据采集处理的复杂度也提高了,从而降低了工程实用性。为了既能够满足工程应用对运算量的限制,又能够保持姿态计算的精度,就需要对旋转矢量的姿态算法进行优化。优化算法的主要内容就是如何在低子样情况下提高姿态计算的精度,因此如何运用陀螺的子样数据就成为优化算法的主要难点。根据对子样的不同运用方式,将旋转矢量优化算法归纳为利用多阶叉乘项和利用前周期采样值的两类算法优化途径。采用双子样采样方式,对上述两类算法表达式进行了推导;采用仿真圆锥运动的方法,分析比较了两类优化算法的误差水平。仿真结果表明,同利用多阶叉乘项优化算法相比,利用前周期采样值的双子样优化算法精度更高,已经达到了传统三子样算法的精度水平,且运算量增加不多,因此上述优化方法具有较较高的工程应用价值。

Rotation vector attitude algorithm plays an important role for decreasing the coning error introduced by the non-commutivity vector. The main method of improving precision of attitude algorithms based on rotation vector is to add the output hits of gym. However, this method can cause great complication of data acquisition, transmission and computing in navigation computer. In order to solve this problem, traditional algorithm should be improved. Dif- ficult point of the improved algorithm is how to use gyro sampling data. Based on the usage of sampling data, two ide- as of improved rotation vector attitude algorithm were proposed in the paper. They are higher order items method and the method which uses gyro output of the last updating period. On the condition of two-sample method, some im- proved rotation vector attitude algorithms were introduced. Analysis and comparison of error level and practicality for these algorithms were presented by simulation method. Result shows that the second method is better than the first method. So the method of using gym output of the last updating period is of great engineering application value.