Error Compensation And Reliability Analysis For Loitering Munition Integrated Navigation System

Loitering munitions are produced as a combination of advanced missiles and unmanned aerial vehicle (UAV) technologies. It aims to fulfilling multiple missions, including cruising round above the target area, investigating, monitoring, damage evaluation, and so on, so as to satisfy the requirement of further information war. The missile-borne navigation system, as the key component of the guidance control system, plays an important role in overall performance of the missile. In order to improve the navigation accuracy and reliability of the loitering munitions system, this thesis focuses on investigating the stochastic modelling and compensation of inertial sensors, fault detection and isolation method, and high-precision integration algorithms. The main contributions of this research are:(1) The Receiver Autonomous Integrity Monitoring (RAIM) for multiple constellations integration system is investigated. In this paper the reliability and separability of measurements from different types of Beidou Satellites (GEO/IGSO/MEO) are compared via simulation data. The performance of integrated systems with multi-constellations is also investigated. It has shown that the Minimal Detectable Bias (MDB) for GEO satellites are the smallest, compared with MEO and IGSO satellites. This is because the measurement accuracy of MEO and IGSO satellites are relative low due to the satellite geometry and signal properties. Furthermore, by fusing multi-constellation GNSS systems the satellite geometric strength is in a large measure enhanced so as to improve the system internal reliability and the FDI capability.(2) The reliability for GNSS/INS integrated navigation system is discussed. The influences of the accuracies of different GNSS systems and inertial systems on the reliability of the integrated system are analysed. Differently with the standalone GNSS system, the measurements’MDBs in the integrated navigation systems are not only influenced by the satellite cut-off angle, residual number and the satellite geometric distribution, but also largely rely on the characteristic of the Kalman filtering model. Besides, high accurate inertial system can help to reduce the measurements’ MDBs in the GNSS/INS integration system, while the improvement is not in direct proportion to the accuracy of the inertial system, due to the interaction between the covariance of measurement noise and the covariance of the state transition noise.(3) Stochastic modelling and compensating for colored noises of inertial sensors are proposed. The navigation accuracy of an inertial system is mainly dependent on the errors of the sensors’ outputs. Therefore, it is necessary to estimate and then to model the error characteristics of the inertial sensors so as to ensure the accuracy and reliability of the navigation system. The bias instability is usually modelled as a Gauss-Markov progress, while it is actually complex and difficult to be determined accurately. In this study, the bias instability processes are theoretically analysed. It has been proved that the bias instability process can be modeled as a sum of independent first-ordered Gauss-Markov processes. The sensors’ errors are then modeled via white noise and random walk, and are used to augment the kalman filter of integrated navigation system. The experimental analyses have demonstrated the efficiency of the proposed method.(4) An enhanced FDI algorithm based on local test of standardised residual vector was proposed to detect and isolate abrupt GPS faults in the integrated navigation system. The influence of correlation coefficient between any two test statistics on the probability of missed detection and wrong exclusion was analysed and used to improve the success rate of fault identification. Real test and simulation data are used to assess the performance of proposed method. The experiment results show that the conventional FDI method cannot always give an optimal solution with the influence of larger correlation coefficient, while the proposed method can effectively improve the reliability of the integrated navigation system by using the validated redundant observation.(5) The high accurate positioning and velocity determination method based on the time differenced carrier phase measurements are investigated for the specific application in loitering munitions navigation systems. With the proposed method, the ambiguity resolution can be avoided so that the computation burden and complexity are significantly reduced. Experiments results have illustrated that the mm/s level velocity can be determined by using the time differenced carrier phases (TDCP). Also, the TDCP method can provide sub-meter level positioning accuracy within 10 minutes of operation time.