Research On Key Technologies Of Single Satellite To Satellite Target Passive Orbit Determination And Tracking

In the recent high-technology local wars,the spacecrafts especially satellites are becoming more and more powerful measures to ensure information superiority as they are not limited by the national boundaries and can travel freely in the global space.With the merits of electromagnetic self-hiding,individual working and easy deploying,the single satellite to satellite passive orbit determination and tracking system has a broad application domain.In a non-cooperative circumstance,this dissertation focuses on the relevant theory problems and key technologies of single satellite to satellite passive orbit determination and tracking,including passive orbit determination and tracking method,system model selection,observability and degree of observability,nonlinear filtering and estimation method and so on.The main contributions of this dissertation are demonstrated as follows:In chapter 2,the kinetic models,in which the target accelerations are decided by their position and velocity vectors,are developed from bearings-only measurements in both two-dimensional and three-dimensional cases.The necessary and sufficient conditions for system observability in the corresponding cases are deduced.The system observability of single satellite to satellite target passive tracking from bearings-only measurements can be fulfilled are then strictly proven in both two-dimensional and three-dimensional cases.Secondly,a satellite to satellite passive tracking method using bearings-only(BO) measurements is put forward.The corresponding extended Kalman filtering(EKF) methods are deduced in detail with the two-body and J₂ perturbation models.Then,the degree of observability with the system and measurement equations in satellite to satellite passive tracking from beatings-only measurements is also addressed.Finally,the estimation accuracy with multiple factors is analyzed through computer simulations compared with the Cramer-Rao lower bounds(CRLB),some instructive conclusions are drawn that can be useful for further research.In chapter 3,a new satellite to satellite passive tracking method using bearings and frequency measurements is proposed in order to counter the defects of the BO method which has long convergence time and large tracking errors.The corresponding EKF methods are deduced in detail with the two-body and J₂ perturbation models,in which the carrier frequency f₀ is divided into two cases:unknown and known to the observer. Then,the degree of observability with the system and measurement equations in satellite to satellite passive tracking from bearings and frequency measurements is investigated as well.Finally,the estimation accuracy is analyzed through computer simulations compared with CRLB.Simulation results show that the passive tracking performance could be efficiently improved while adding frequency measurement into bearings-only system when the two satellites are not in the same orbits,which presents a new system for satellite to satellite passive tracking.In chapter 4,the particle swarm optimization(PSO) is introduced into satellite to satellite passive orbit determination and a new satellite to satellite passive orbit determination algorithm from frequency-only(FO) measurements is proposed.The state estimation method is deduced based on PSO with the two-body and J₂ perturbation models,in which the carrier frequency f₀ is also divided into two cases:unknown and known to the observer.Then,the degree of observability with the orbital elements of the target satellite under certain geometry relationship in satellite to satellite passive orbit determination from frequency-only measurements is studied and probed.Next,the CRLB is deduced in detail.Finally,the estimation precision is analyzed through computer simulations compared with CRLB.Simulations indicate that the proposed method based on PSO can solve the passive orbit determination problems with the two-body and perturbation models commendably.Moreover,the proposed method is also superior in computation complexity than the grid search method.In chapter 5,the nonlinear filtering methods are investigated to improve the estimation performance based on the passive tracking methods proposed in chapter 2 and chapter 3 with the perturbation models.To counter the weak observability problem of the BO method in the Cartesian coordinates system,which is caused by the nonlinear measurement equation,a satellite to satellite passive tracking method in the modified spherical coordinates(MSC) system is proposed.Regarding the problems that the system model error changes from time to time,the adaptive fading extended Kalman filtering(AFEKF) method is investigated.To overcome the first order approximation of EKF with complex system model,a satellite to satellite passive tracking method based on unscented Kalman filter(UKF) is also derived.Simulation results show that among the three modified nonlinear methods,the estimation accuracy of UKF is the best, followed by AFEKF and MSC method,and all the three methods are effective in terms of estimation quality than the EKF methods proposed in chapter 2 and chapter 3 with the perturbation models.The ephemeris data in chapter 4 is propagated by the fourth order Runge-Kutta method,the other ephemeris data in the residual chapters are all generated by the general satellite simulation software STK6.0.