几类捷联惯导机械编排方法及其运动学约束

SEVERAL SINS MECHANIZATION METHODS AND KINEMATIC CONSTRAINTS

针对低精度惯导解算累积误差增长迅速的缺点,提出利用速度约束进行阻尼,并给出了相应的Kalman滤波观测方程。实验结果表明,对于高精度数据,泰勒展开法精度较高,对于低精度数据,速度约束极大地延迟了可接受误差范围内的导航时间。

The differences among SINS mechanization methods of rotation vector,quaternion and Taylor expansion were analyzed,the results indicate that the posture expression and the appropriation degree of algorithm are main reasons caused different calculation precision in those methods. A velocity constraints method and corresponding Kalman filter model were proposed. Test results show that higher accuracy can be obtained by Taylor expansion method for high accuracy analog data,and greatly delay navigation time cab be delayed by velocity constraint method for low accuracy analog data.