Research On Estimate Algorithm Of Boost Phase Trajectory Based On Space-Based Early Warning System

Space-based infrared early warning system is an important part of missile defense system. The estimate algorithm of missile's states and trajectory's parameters by early warning satellite is a key technology of information processing. In this dissertation, missile's states in the boost phase and trajectory parameters estimate algorithms are studied by using angles of arrival measurement from space-based infrared sensor, and the availability and precision is validated through simulation.According to the character of ballistic missile in boost phase, the gravity turn model is established, and according to the position relationship between early warning satellite and the target, the measure model is established.Aiming at the nonlinearity of state and measure equation and measurement being only angle information, which results in poor observability and classical estimate methods such as extended Kalman filter not converging, the boost phase states and covariance are estimated using unscented Kalman filter(UKF). Simulation shows that UKF algorithm can estimate missile trajectory with satisfying precision and convergence speed, the method's availability and feasibility for missile estimation is validated. The effect of measure noise on the estimation precision is analyzed, which provide reference for more accuracy estimation.As the measure noise has much effect on estimate precision and convergence speed, when noise increases abruptly, the position estimation converges slowly and the precision becomes poor. Aiming at this problem, particle filter is studied. Combination filter algorithm is constituted using UKF and improved particle filter (UPF), and missile's states with noise increase are estimated. Combination algorithm fully use the advantage of the UKF for small noise and the UPF's non-sensitivity to noise, when noise increase, UPF can also estimate the missile's states well.The missile's lunch point parameters and burnout point parameters are estimated using boost phase state estimation, and the impact point is predicted according to ellipse trajectory theory. And the effect of the estimation error of the burnout point on the impact estimate precision is analyzed. Simulation shows that the impact point precision is mainly dependant on the estimate precision of the burnout point's velocity.