The Design Of SRS/SINS/CNS Auto-Nomous Navigation System And The Extension Research Of CKF Algorithm

High-precision autonomous navigation system and high-performance filtering algorithm are not only an important guarantee for the accurate navigation of modern aircraft,but also an important index of national defense high-tech development level.However,the existing navigation systems still have certain shortcomings in autonomy,reliability and filtering solution accuracy.Thus,it is necessary to conduct further research and improvement to meet the demands of Chinese national defense construction for precise navigation of modern aircraft.As an innovative concept of navigation,the spectral red-shift autonomous navigation not only has high navigation accuracy and strong autonomy,but it also has great real-time performance,leading to a new technique to improve the autonomy of spacecraft navigation,expand navigation strategies,and realize the autonomous operation of spacecraft.As a result,this method has attracted great attention in the navigation field.This paper studies the key issues in spectral red shift autonomous navigation,high-precision Strap-down Inertial Navigation(SINS)/Spectral Red-shift(SRS)autonomous navigation system design,SINS/SRS/Astronomy(CNS)multi-source fusion autonomous navigation system design,integrated navigation nonlinear high-performance filtering solution method,integrated navigation multi-source information fusion,and system noises estimation and errors compensation.The main research results and innovative contributions are reflected in the following aspects:(1)Based on the solar system celestial spectrum red-shift speed navigation principle,using the spectral red-shift information of the natural celestial bodies of the solar system and the law of inertial motion of objects,a new high-precision SINS/SRS autonomous integrated navigation system is developed.The principle,scheme and mathematical model of SINS/SRS autonomous integrated navigation system are established,and simulation and experiments are carried out for performance evaluation of the above model,scheme and system design.(2)Based on the spectral red-shift information of the natural celestial bodies of the solar system,the law of object motion and the astronomical information of natural celestial bodies as navigation information sources,a high-precision SINS/SRS/Astronomy(CNS)multi-source fusion autonomous integrated navigation new system is developed by using a nonlinear filtering algorithm and multi-source information fusion technology.The principle,scheme and mathematical model of SINS/SRS/CNS multi-source fusion autonomous integrated navigation system are established.A robust adaptive Unscented particle filter algorithm suitable for the filtering calculation of this integrated navigation system is also developed.The subsystems of SINS/SRS and SINS/CNS,and the SINS/SRS/CNS multi-source fusion autonomous integrated navigation system are established respectively.Simulations and experiments are conducted to verify and validate the principles,schemes,models and algorithms for SINS/SRS/CNS integrated navigation system.(3)In order to overcome the shortcomings of the Cubature Kalman filter algorithm,a random weighted cubature Kalman filter(RWCKF)algorithm is proposed.In this algorithm,according to the different size of estimation error of each Cubature point by using random weighted estimation method to adjust the estimated weight.Different weights are allocated to each cubature point to estimate the state prediction and measurement prediction as well as their error covariance.The disturbances of status prediction error,measurement prediction error and their error covariance on the filtering accuracy are inhibited effectively,leading to improved filtering accuracy.The random weighted cubature Kalman filter estimation models are established for state prediction,measurement prediction and their covariance,auto-covariance and cross-covariance of nonlinear system.Finally,the proposed random weighted cubature Kalman filter was applied to target tracking system and SINS/SRS autonomous integrated navigation system for performance evaluation and analysis.The validity and superiority of the proposed random weighted cubature Kalman filter algorithm are proved.(4)Aiming at the filtering calculation problem of nonlinear systems with constant noise,an adaptive random weighted cubature kalman filter algorithm is presented.The random weighted estimation model is established for nonlinear system state prediction,measurement prediction and corresponding covariance matrix.This method combines adaptive filtering and random weighting to improve the cubature Kalman filter(CKF)performance.It adaptively adjusts the weights of cubature points to enhance the prediction accuracy and inhibit the disturbances of system noises and measurement noise on state estimation,leading to improved filtering accuracy and robustness of nonlinear system.Finally,the comparison analysis of the standard CKF and the proposed ARWCKF is conducted through simulations,proving that the proposed ARWCKF has a better filtering performance than CKF and randomly weighted CKF.(5)The random weighted estimation model is established for nonlinear systems with constant noise.It is proved that the random weighted estimation is unbiased for system process noise mean and measurement noise mean,while the variances of system process noise and measurement noise are biased.Techniques are further developed to compensate for the errors involved in variances of system process noise and measurement noise,and the accuracy of noise estimation is verified and evaluated.In this paper,the research outcomes and achievements have an important effect on improving the navigation accuracy and filtering performance.They also contribute to integrated navigation system design,nonlinear system filtering calculation,multi-source information fusion and integrated navigation system error estimation and error compensation.This research results can not only be used to improve the accuracy of navigation and calculation of the spacecraft,but they can also be applied to improve the accuracy of navigation and positioning of vehicle in the fields of aviation,sailing and transportation.