Research On Some Key Technologies Of MEMS-INS/GNSS Integrated Navigation System With The Application To Guidance Munition

The guidance of the conventional ammunitions is an important way to improve the attack accuracy.The conventional ammunition has the characteristics of both high rotation and flight speed,and the key technology to guarantee the attack accuracy of the conventional ammunitions is to obtain the real motion information of the ammunitions accurately.Thus,adaptive and robust integrated navigation technology appears as the focus in the research field of navigation with the application to ammunitions,which is of great importance to the guidance of the conventional ammunitions.Aiming at making full use of measurement information from each of the sub-navigation systems to improve the accuracy,reliability and suitability performance of the ammunition-borne MEMS-INS/GNSS integrated navigation system,this dissertation focuses on the key technology of developing an ammunition-borne integration navigation platform based on the research of the advanced research projects in the early stage.The algorithm and scheme of the integrated navigation system with high accuracy and reliability that is suitable for the complex ammunition-borne environment is proposed.Based on the extensive investigation of MEMS-INS/GNSS integrated navigation system,the research methods and difficulties of the MEMS-INS/GNSS integrated navigation technology with application to the guided ammunitions for both at home and abroad are compared and analysised.And from which,the research contents and technical routs are determined in this dissertation.The main contents of the study include the following:(1)Aiming at the problem that the process uncertainty of the MEMS-INS/GNSS integration navigation system will occur in the extend range stage,the impact of process uncertainty induced by the engine's vibration on the precision performance of integration navigation system is analysised and a process uncertainty suppression algorithm based on the improved strong tracking filtering technology is proposed in this dissertation.In the proposed algorithm,the adaptive tracking for the corresponding navigation parameter is achieved by introducing the multiple suboptimal fading factor into the prediction covariance matrix;the precision loss in the free flight stage is avoided by employing the hypothesis test method to identify the process uncertainty.On this basis,a new high degree spherical simplex-radial cubature rule is derived,which further improves the estimation accuracy of the traditional strong tracking filtering and finally achieves the effectively process uncertainty suppression.(2)Aiming at the problem that the performance of the MEMS-INS/GNSS integration navigation system may degrade when the GNSS measurements is contaminated by the heavy-tailed non-Gaussian noise,the degradation mechanism of conventional integrated filtering on the precision performance under the condition of non-Gaussian measurement noise is analysised and an adaptive robust filtering algorithm based on the newly constructed adaptive maximum correntropy criterion is proposed in this dissertation.In the proposed adaptive robust filtering,the adaptive maximum correntropy criterion is firstly constructed,and based on which,the adaptive robust filtering is then designed.Also,the methods of how to determine the optimal parameters in the proposed filtering are respectively discussed.Besides,the proposed filter is further optimized from the view of numerical problem avoiding and computational complex reduction.Finally,the adaptive robust state estimation is achieved for the integrated filtering with non-Gaussian measurement noise,which greatly improves the accuracy of the navigation parameter for the MEMS-INS/GNSS integrated navigation system under the condition of GNSS measurement outliers.(3)Aiming at the problem that the MEMS-INS/GNSS integrated navigation system's powering-on before the artilery launching and the ammunition's arbitrary maneuvering after the artilery launching are both unfeasible in practical application,an in-flight alignment method aided by GNSS measurements is investigated and an close-loop dynamic attitude estimation based in-flight alignment scheme using multiple vector observation in inertial reference frame is proposed.The mathematical model of dynamic attitude estimation technology is firstly constructed and the optimal resolving procedure of the multiple vectors to determine the attitude is derived.The real-time drift of gyroscope is then introduced in the process of solving the attitude determination vector by using closed-loop feedback strategy,by which the high accuracy of initial alignment is ensured.Additionally,the quaternion-based multi-vector attitude determination algorithm is adopted to improve the convergence speed of the algorithm,which further ensures the high speed of initial alignment.Finally,the initial parameter acquisition with both fast and high precision is achieved by the MEMS-INS/GNSS integrated navigation system with application to the guided ammunitions.(4)Based on the above theoretical study,the effectiveness and superiority of the proposed algorithm for each key technology are respectively validated by the Monte Carlo simulations and car-mounted experiments.Finally,the flight tests are conducted to make a comprehensive assessment and validation for the above key technologies.The research works in this dissertation is of great referential values for the MEMS-INS/GNSS integrated navigation system in the application of guided ammunition.