Research On Control Methods Of Inner-formation Gravity Field Measurement Satellite System

The Inner-formation gravity field measurement satellite system (IFS) is a novelprogram aiming to measure the Earth’s gravity field. It consists of the outer-satellite andthe inner-satellite and hereby turns out to be a compact formation. Completely enclosedin the outer-satellite, the inner-satellite is shielded from non-conservative forces andfollows a purely gravitational orbit. With measurement of the orbit, the Earth’s gravityfield can be reconstructed. One of the key techniques determining the measurementlevel of gravity field is the outer-satellite control used for keeping the inner-satelliteflying in a pure gravitational orbit stably. To resolve the control problem for IFS, thisdissertation studies the integrated relative orbit maintenance and attitude stabilizationcontrol, which involves dynamics modeling and analysis along with control accuracyrequirements and analysis, and proposes constrained nonlinear model predictive control(CNMPC) method, disturbance compensation control and all-propulsion control scheme.The main issues and results of this dissertation are summarized as follows.The relative orbit and attitude dynamic models are constructed and thedisturbance environment is analyzed for IFS in steady-state phase. Based onanalysis of short-term and long-term effects of main disturbances on the relative orbitand attitude, drag compensation is established as the primary control object to maintainthe relative orbit.The accuracy requirements of relative orbit maintenance and attitudestabilization control of IFS are specified. Based on the principles of gravity fieldmeasurement and relative position measurement of IFS, the relationships betweenrelative orbit maintenance and attitude stabilization control and gravity fieldmeasurement, relative position measurement and POD are formulated. The accuracyrequirements of control system are analyzed to provide specification for the controlsystem design, using theoretical modeling, error propagation analysis and simulationexperiments.Three techniques are developed for the relative orbit maintenance andattitude stabilization control, and the CNMPC method is proposed to implementthe integrated orbit and attitude control for IFS. An adaptive full-state feedback lawis taken to account for the nonlinearity and parameter uncertainty behaved by both therelative orbit and attitude dynamics. To improve control accuracy, shorten regulationtime and remove input chattering, a general nonlinear model predictive controlalgorithm is developed to deal with multi-input-multi-output system, achievingrobustness against the model parameter uncertainty and exogenous disturbance. Takingthe territory of inner-satellite and performance restrictions of the actuator into account,the CNMPC algorithm is proposed. The invariance with respect to uncertainties and disturbances applies to the mission requirements of IFS, namely the strong constraintson relative position and high attitude stabilization.A disturbance compensation scheme based on UKF (Unscented Kalman Filter)and disturbance estimation is put forward. To achieve disturbance rejection for therelative orbit maintenance, a real-time estimation of disturbance using UKF is projectedbased on observation of relative position and attitude. Making use of the estimationresults, the disturbance compensation is realized by means of disturbance estimation andfeedback, improving the control performance of relative orbit maintenance.An all-propulsion design using continuous micro-thruster to perform theintegrated orbit and attitude control of IFS is brought forward. The all-propulsionscheme is realized by a thruster configuration adapting to playing drag compensationand balance control simultaneously.This dissertation aims to figure out the precision control problem of IFS and hasimportance theoretical significance and engineering values in improving the controlperformance of IFS, embodied by proposing the CNMPC algorithm and the real-timeestimation method of disturbances using UKF. It can also provide theoretical andtechnical references on the control system design for satellites.