SINS/GPS/OD Fault-tolerant Integrated Navigation System

The single navigation system is susceptible to various environmental disturbances in actual work and cannot meet the high-precision,high-reliability navigation requirements.Based on the navigation system such as SINS,GPS and OD,this paper studies the faulttolerant navigation system composed of three,aiming at improving the accuracy and reliability of the integrated navigation system,and ensuring that it works properly in a dynamic environment.In this paper,the federal filtering scheme of SINS/GPS/OD integrated navigation system is studied.The error of the lever arm existing in the integrated navigation process is analyzed.The lever arm error on the positioning result is added to the state quantity to establish the system equations and measurement of the SINS/GPS and SINS/OD subsystems respectively.The equation compensates for the lever arm error and the like by filtering estimation.Aiming at the shortcomings of the traditional fixed information distribution scheme,an adaptive information distribution method is proposed.The simulation results show that the positioning accuracy of the federated filtering method for adaptive information distribution is better than the traditional method.The fault detection algorithm of SINS/GPS/OD integrated navigation system is studied.For fault detection of sudden faults and gradual faults,in order to compensate for the deficiencies of different detection methods,the residual detection and SPRT detection are used to perform single-component joint detection on each component of the residual to adapt to the tracking ability of the system.Simulations show that single-component joint detection improves the sensitivity of detection compared to a single detection method.A fault-tolerant algorithm for adaptive adjustment is proposed.Since the local filter estimation of the sub-filter still enters the main filter for information fusion during the period from the occurrence of the slow-change fault to the detected fault,based on the singlecomponent joint detection method,the quality is jointly designed by the two detection methods.The adjustment factor quantifies the influence of the current fault on the sub-filter,and adaptively adjusts the weight of the sub-filter in the global estimation,so that the global fusion obtains the optimal estimation.Simulation results show that this adaptively adjusted fault-tolerant algorithm improves the fault tolerance of the entire navigation system.