基于容错惯性网络的相对导航方法

In order to meet the requirements of accurate detection and reconnaissance of external environment, the advanced multi-load general-purpose aircrafts usually carry multiple mission equipment. Aiming at the problem of wing deflection, low precision and random failure in sub-IMUs, a relative navigation method based on fault-tolerant inertial network is proposed. Firstly, the dynamic relationship conversion model between multi-node IMUs under flexural deformation is established to form redundant measurement information, and the least squares fusion based on generalized likelihood ratio detection is performed to improve the reliability and accuracy of measurement data. Secondly, the local motion information between the inertial networks is used for the calculation of relative navigation and the error estimation and compensation models are established. Simulation results show that the proposed method can effectively complete the high precision relative motion estimation between the main node and the slave nodes in the case of random faults of the inertial sensor, and the relative position estimation errors of the three directions are 0.0646 mm, 0.0634 mm and 0.7377 mm, respectively.