Research On Nonlinear Integrated Navigation Technology Of Redundant MEMS-SINS

In order to ensure that the navigation system can work stably for a long time,redundant methods can be used to improve the reliability of the Strapdown Inertial Navigation System(SINS).The small size of the Micro Inertial Measurement Unit(MIMU)provide conditions for the application of redundant technology.In order to make up for the shortcomings of SINS error accumulation over time and the error problem of low-cost MEMS inertial devices,the characteristics of the redundant MEMS-SINS and the Global Navigation Satellite System(GNSS)that can compensate for each other in terms of error propagation performance can be integrated.However,on the one hand,it is necessary to further research how to suppress the influence of low accuracy of MEMS inertial devices on the attitude angle estimation of integrated navigation.On the other hand,it is necessary to further research how to deal with the problem of interference and failure of GNSS signals.Information fusion technology is also an important factor affecting the effectiveness of integrated navigation.How to make better use of information fusion to improve estimation accuracy needs further research.In order to ensure the reliability and navigation effect of the MEMS integrated navigation system,this article will conduct research in the following aspects:Firstly,choose the device-level redundancy method to improve the reliability of the system,determine the number and structure of the system's redundant configuration by performing a reliability analysis on the system.And consider how to perform system reconstruction after the inertial device fails.The nonlinear filtering method of integrated navigation system is introduced and analyzed,and the Unscented Kalman Filter(UKF)algorithm suitable for strong nonlinear conditions is introduced.Aiming at the problem of GNSS signal failure,after the failure,the OD and redundant MEMS-SINS are used to ensure the consistency of the navigation effect of the system.The integrated scheme of redundant MEMS-SINS/GNSS and redundant MEMS-SINS/OD is designed and completed,and the switching rules of working modes are given.Secondly,because the observability of the system state quantity can reflect the effect of state estimation,for the divergence problem of the course angle estimation,the specific reason of the observability analysis of the nonlinear system can be find out.Because the observability analysis method of nonlinear systems is not uniformly defined in academia at present,in order to be able to analyze it quickly and easily,Taylor series expansion can be used to approximate it to linear time in a period of time.Then,use the PWCS and SVD methods to perform observable qualitative and quantitative analysis on the state quantities of the redundant MEMS-SINS/GNSS.Thirdly,in view of the problem of weak observability of the course angle state,the course constraint of the system by the track angle obtained from the single baseline GNSS measurement,which effectively improves the estimation effect of the system's course angle.Aiming at the problem of the sideslip angle of the track angle measured by the single-antenna GNSS during the turning state,the movement state of the system is judged by the turning judgment rule.When the system is detected in the turning state,the course constraint is turned off,and only use MEMS for achieve short-term effective output.Aiming at the influence of the low precision of MEMS devices and turning off the course constraint on the system navigation effect,the nonlinear noise statistical estimator based on Sage-Husa and the adaptive UKF algorithm improved by fuzzy reasoning were used to suppress this effect in a short time.The results show that the algorithm can effectively ensure the system navigation accuracy.Fourthly,for the problem of GNSS signal failure easily,when the GNSS signal failure is detected,the OD and the redundant MEMS-SINS are used to ensure the consistency of the system's navigation effect.Aiming at the problem that the mileage measurement information is easily disturbed by outliers,first analyze the outliers that may appear in the mileage measurement information,and then outlier detection rules are established.When the system detects outliers in the measurement information,the conventional UKF algorithm is improved by the innovation vector to complete the design of an improved anti-outlier adaptive UKF algorithm.Simulation experiments prove that this method can effectively improve the influence of Odometer outlier information on the system's navigation effect.Finally,the filters of the redundant MEMS-SINS/GNSS part and the redundant MEMS-SINS/OD part are used as sub-filters 1 and the sub-filters 2.The federal filter method is used to combine the two sub-filter filtering methods to verify the navigation effect of the overall integrated navigation system.The simulation verification proves that the overall scheme can achieve continuous,high-precision,high-reliability navigation purposes.