机载关键位置间相对姿态获取受机体挠曲影响显著。提出一种基于分布式惯性测量单元(IMU)的相对姿态估计方法。首先给出了IMU间相对姿态矩阵微分方程,计算节点间相对姿态,然后对局部区域挠曲形状拟合。利用相对姿态计算值和拟合值构造量测量,结合相对姿态误差方程,通过Kalman滤波器估计相对姿态误差。仿真结果表明,各节点用1°/h陀螺时,相对姿态估计精度优于8′。该方法可为机载关键位置相对姿态精确获取提供技术参考,尤其当系统无高精度惯导且变形统计特性未知时。

The flexure during flight adversely affects determination of relative attitude between aircraft's critical local frames. A relative attitude estimation method was presented based on information from distributed inertial measurement units(IMUs). Firstly, a differential equation of relative attitude matrix was derived so that a calculated relative attitude between distributed nodes was obtained. Then a fitted value of relative attitude was achieved with fit flexure shape of local area. By utilizing the calculated value and fitted value mentioned above and derived relative attitude error equation, a state-space model was given and a Kalman filter was used to estimate the error therefore. Simulation results indicate that when the gyroscopes' precision of distributed IMUs is 1º/h, relative attitude estimation accuracy reaches to 8. In this paper, a technical reference is illustrated to address accurate acquisition of relative attitude between nodes especially when the IMUs all are of low precision and statistical characteristics of flexure is unknown.