Research On Attitude Control Technology For Video Satellites Tracking Ground Targets

Micro/Small video satellites are developing rapidly in recent years and abundant achievements have been made,on which universities,research institutes and capital market attach much importance.It shows the special status of video satellites due to the need of society.This article starts with staring control problem,then conducts research and analysis on control algorithm,modeling and application scenarios,aiming to further dig out the potential of video satellites.Following contents are researched:First the framework of staring control of video satellites is established.For video satellites,staring control is the basic work mode during Earth observation.The attitude tracking equation is rooted in attitude dynamics equation of solid body,and ellipsoidal model is introduced to simulate a ground target,thus the basic ground staring control problem is realized.The duration,when the target is observable for the video satellite,is limited and precious,thus to realize rapid maneuver is rather significant.The time-optimal attitude maneuver problem is solved by Legendre pseudo-spectral method in this dissertation.Then considering nonlinearity of attitude dynamics equations and limitation of angular rates and maximum torque,the paper builds Bolza-form time-optimal control problem and converts the former problem into a nonlinear programming problem(NLP)with interpolating globally in LGL points.On contrast,direct shooting method which only disperses the control torques is introduced in order to analyze the efficiency and effectiveness of LPM.The result demonstrates that LPM is of high accuracy and can meanwhile lessen the calculation time compared with direct shooting method.Normally reaction flywheels are adopted as the attitude actuators of micro/small video satellites,thus the performance of flywheel is able to directly affect the accuracy of stable staring control.Experiment on a flywheel with the rated torque output 0.1Nm is conducted,and response data of specified torque order and flywheel rotate-speed are recorded.Combining the theoretical model and the recorded data,it establishes the practical models of friction torque and transfer function of torque mode.This model is introduced to staring control for the purpose of comparing the impact on the stable attitude control of this newly established flywheel model and the ideal model.Currently the video satellite mainly observes point target,only covering the area that is included in a single field of view of the camera,which greatly limits its observation capacity.However,domestically the area observing for a video satellite remains not explored academically,while the area target observation is profound in practice.To analyze and promote the area target observing ability,a method based on the feedback of real-time covered area and dynamical strip decomposition is proposed.This method requires calculating the real-time area scale that is covered by the field of view of the camera,and then through attitude maneuver,the camera points to the next field of view according to certain schedule.When it finishes observing a strip of the area target,the method decides dynamically the scale of the next strip.This method has adequately considered the complexity and irregularity of the area shape covered by camera,and tries to avoid the overlapping of two fields of view,therefore it can improves the observing efficiency and covers a larger scale of area target.