| As China's self-built satellite navigation system Beidou accelerates its ultimate goal of global service,it has gradually deepened research on constellation composition,signal transmission,and ground enhancement,and the overall system performance has been continuously improved.However,subject to the principle that satellite navigation systems are susceptible to interference,a single Bei Dou system cannot adapt to some special occasions where signal occlusion is severe or carrier mobility is too strong.Therefore,the INS/GNSS deeply-coupled technology has gradually attracted the attention and discussion of many domestic research institutions in recent years,because it has special performance advantages in anti-jamming and high-dynamic applications.And the research on aspects of signal processing,loop architecture,fusion filtering,and hardware implementation have become an important direction in the field of navigation.Solving the key issues in the above-mentioned aspects has important theoretical reference significance and value for the performance improvement and engineering application of deeply coupled systems.This paper focuses on the key problems of INS/Beidou deeply coupled system under high dynamic environment.From the aspects of fusion precision and fault tolerance reliability,the research on performance enhancement technology of deeply coupled system is carried out.The Beidou heterogeneous constellation characteristics and the deeply coupled system loop model are analyzed.Based on this,the deeply coupled system stability judgment,fusion filtering,and fault detection and fault tolerance methods are studied.Finally,a semi-physical simulation experimental platform has been established for theoretical verification and the prototype system is developed.The main research content includes the following aspects:Firstly,the modeling of INS/Beidou deeply coupled system and characteristics analysis are carried out.On this basis,the Beidou positioning stars optimization algorithm was proposed and analyzed.The computational efficiency of the method is superior and the correctness of the method is verified.Then the frequency domain analysis method is used to analyze the loop structure of the system.And the principle and effect of inertial assistance are analyzed.What's more,the nonlinear characteristics of the receiver loop and inertial navigation system under high dynamic conditions are studied which can provide a theoretical basis for the subsequent increase in the performance of deeply coupled systems.Secondly,the stability judgment algorithm of system loop is studied for the parameter configuration of the system.The factors affecting the stability of deeply coupled loops under high dynamic conditions are analyzed.For the requirements of the precision for inertial devices,an analysis method of the inertial error propagation characteristics based on norm analysis is proposed.And a complete stability judgment for deeply coupled system is given.It can effectively judge the loop stability of the system parameters under the specified dynamic index and greatly improve the calculation efficiency.Then,in order to improve the navigation accuracy of the system further,a high-dynamic INS/Beidou deeply coupled navigation filter fusion technology is studied.Firstly,an intergrated algorithm based on classification measurement is proposed to make full use of the characteristics of the Beidou constellation.In order to improve the filtering accuracy,nonlinear loop pre-filtering algorithm and combined filtering algorithm based on IQ information are studied,and a scheme for generating deeply coupled loop control signals is proposed.Experiments show that the proposed algorithm can effectively improve the loop tracking accuracy and navigation performance of the system under high dynamic environment.Besides,in order to improve the reliability of the system,a high-dynamic INS/Beidou deeply coupled navigation fault tolerance technology is studied.Firstly,the advantages and disadvantages of time series analysis methods and artificial neural network analysis methods are discussed.An inertial-assisted loop fast fault detection algorithm based on support vector regression is proposed in the framework of deeply combination.Then a subset detection-based fault detection and system reconstruction method is proposed for multi-star faults,enabling the system to using fewer visible stars to determine satellite faults.Experiments show that this method can effectively improve the fault recognition rate of deeply coupled system under high dynamic conditions,and mitigate the impact of high dynamic on navigation results.Finally,an INS/Beidou deeply coupled multi-source fault-tolerant navigation simulation test platform is designed.The hardware test system based on FPGA and DSP and a software simulation platform based on Matlab are performed.And the test verification system proposes an ARM-based inertial information real-time synchronization scheme.The performance of various algorithms proposed in this paper is tested and verified,indicating the performance enhancement effect under high dynamic conditions.This paper focuses on deeply coupled system performance enhancement related technologies in high dynamic environment,and carries out detailed and in-depth exploration and research work on deeply-coupled loop theory and hardware implementation,which lays the foundation for further deepening the research of deeply-coupled technology. |
PDF Full Text Request