| The distributed InSAR satellite system equips SAR antennas on close formation flight satellites, and completes SAR interferometry mission through the collaboration of small satellites and SAR antennas. It is a kind of new conceptual radar system with enormous potential, which faces many challenges in basic theory and technology. The high precision determination of the formation spatial states is one of the key technologies, and is the important guarantee for the success of the distributed InSAR mission. It is also a common problem to solve in other satellite formation flight missions. On the background of the distributed InSAR satellite system, this dissertation clarifies the relation between system and spatial state measurement, emphasizing on the two different measurements dual frequency GPS and inter-satellite wireless, and studies the high precision spatial state determination of satellite formation. The main work includes five parts:Firstly, the spatial geometrical relation of distributed InSAR geolocation is analyzed, the relation between system parameters and spatial states measurement is explained, and the significance of spatial states measurement is illustrated through requirement analysis of the distributed InSAR satellite system. Closed-form solution to the geolocation is provided and sensitivity of the solution is analyzed. The formulations of geolocation precision criterion and system parameters precision criterion are proposed. Interferometry baseline is defined. Interferometry baseline is involved with spatial baseline and time baseline. Time baseline is obtained by master satellite absolute orbit determination, and spatial baseline is obtained by high precision inter-satellite relative position determination. The transfer from measured baseline to spatial baseline needs body correction and time synchronization. The precision analysis of body correction is provided.Secondly, the pre-processing method of onboard dual frequency GPS observations is studied, and data pre-processing scheme and algorithm is proposed. A novel robust Vondrak filter is presented, which can ensure that the gross errors are not too influential on the curve fit, and makes the original model robust, when fitting function of the original signals is unknown and they are contaminated by gross errors. The quality evaluation of onboard dual frequency kinematic GPS observations is studied. Computational experiment is carried on to test the correct and validity of the algorithm by pre-processing and evaluating the onboard observations of CHAMP satellite.Thirdly, Desampling method of onboard dual frequency GPS observations is studied. The advantage and disadvantage between polynomial smoothing pseudo-range method and carrier phase smoothing pseudo-range method are analyzed. Zero-differential kinematic orbit determination of onboard dual frequency GPS is studied. Based on quality analysis of GPS observations, a reasonable weighed method of code and phase is presented. Zero-differential dynamic orbit determination of onboard dual frequency GPS is studied. Observations on CHAMP satellite for one week are tested. Zero-differential dynamic orbit determination results are compared with GFZ science orbit. The RMS (root of mean square) in R, T and N component are 4.8cm, 8.2cm and 7.5cm, and the RMS in three-dimension is 12.2cm. Precise orbit evaluation with satellite laser ranging (SLR) data is studied, based on which a new method to calibrate orbit system error is proposed. The new method gets SLR data from several stations in one pass arch, makes orbit error projection at different time, and estimates orbit system error.Fourthly, High precision determination of relative position based on CDGPS is studied. Projection model of single-layer ionosphere path delay for onboard GPS is constructed. Simulation based on the model analyzes how the differential ionosphere delay affects the relative position determination of distributed InSAR satellite formation. A CDGPS relative position method of distributed InSAR satellite formation based on dynamic orbit model is presented. Based on CDGPS measurement and the prior restricted information provided by dynamic orbit model, the observation of long arch is computed, which controls the random errors in the measurement efficiently, improves the relative position determination precision, and comes over the disadvantage that kinematic single point position determination can not be used when observation geometry is bad or observations are not enough. The contribution of inter-satellite distance measurement to CDGPS relative orbit determination precision is analyzed.Finally, High precision relative states determination based on inter-satellite wireless measurement is studied. The principle of inter-satellite wireless measurement is introduced. The advantage and disadvantage of several different traditional relative state estimation methods are compared. A new inter-satellite relative state determination method based on spline model is proposed. The method transfers the direct estimation of state parameters in traditional method into spline parameters, reduces the number of parameters to estimate, improves the sensitivity of distance change to the state parameters, and makes the estimation structure more stable. The signal coverage and shelter of inter-satellite measurement and the geometric effect of formation design to wireless measurement is discussed.
|
PDF Full Text Request