Thrust Dynamic Allocation Of Spacecraft With Redundant Thrusters

Thrusters are frequently applied to orbital and attitude control in vast spacecraft missions among many kinds of actuators. To improve the ability of maneuver and on-orbit stability, high-performance spacecraft often deploy abundant thrusters that are larger than the number of system freedoms. Thus, it makes the allocation procedure, projecting the desired control commands to thrusters’ allocated commands, not unique. At this back ground, control allocation technique is proposed as an effective solution by dynamic alloction of thrusters on over-actuated spacecraft. This thesis is funded by Doctoral Program of the Ministry of Education Project, “Research of dynamic control alloction in spacecraft and spacecraft foramtion”, and Technology and Industry for National Defense Civil Astronautic “Twelfth-five” Advanced Research, “xxx modularization satellite plateform technic research”. The thesis studies problems of orbit prior control in orbit-attitude integrated control, fuel balancing in formation and single spacecraft, and robust control by making use of design freedoms in over-actuated thrusters and control allocation. In the research, the thesis proposes several kinds of control allocation algorithms to meet the different requirements. The research achievements are listed as follows.On-orbit dynamic allocation among thrusters is a key factor impacting system performance of a spacecraft. Traditional fixed allocated algorithms usually need a vast storage of date and do not give a solution when thrusters encounter failure. Hence, we propose a convex cone based and mixed offline/online look-up table method in over-actuated spacecraft. In addition, we explain how to modify the algorithm on line. The algorithm utilizes all possible thrusters’ conbinations, which are decided by vertex-expended convex cone, chooses thrusters conbinations online and determines the on-off options. When thruster fail happens, we adjust the convex cone to isolate the failure thrusters. The proposed method ultilizes convex analysis to schedule the look-up table,and makes online table more effective under complicated configuration. In this way, online calculation is significantly reduced and thruster failure can be properly handled. Finally, we compare the proposed method to fuel minimum method and algebraic look-up table algorithm. Under almost the same effectiveness of fuel consumption, the simulation results show the feasibility and effectiveness with reducing the average error by 26.85%.As a kind of common actuators, thrusters are not only used in orbit control but also in attitude control. Therefore, how to ultilize them to control the orbit and attide is a big problem in high performance spacecraft. In order to reduce fuel consumption, we bring up the idea to first satisfy the orbit control requirements without consuming more fuel in orbit-attitude integrated control. Based on attainable-set modified factors technique, we design a convex slack optimization method and dimension-reducing optimization method in different control effectiveness matrix and thrusters configurations. Under the premise of ensureing thruster requirements, torque requirements are met in the largest extent to reduce the total fuel consumption. The simulation explains the effectiveness and feasibility of proposed algorithm.Thruster’s uncertainties will cause additional fuel consumption, change orbit and attitude status, prolong system stable time, reduce control performance, and so on. So it is important to make full use of additional system freedom providing by redundant thrusters. Minimum singular value depicts the robust performance of redundant deployed configuration. The thesis studies system robustness under the possibilities of redundant deployed configuration in a quantitative way, and describe the uncertainty model as ucertainty ellipsoid set. In this way, the uncertain parameter problem can be transferred to a fixed optimalization model. The proposed optimal spacecraft robust allocation algorithm can optimalize the allocation effectively and improve the robust performance due to different uncertainties. Simulation results show a reduction of total and maximum allocation error by 15% when there is a random 5%-20% error in control matrix.Thrusters are also key factors in realization of spacecraft formation intergrated missions. The lifetime of the entire formation is decided by the inner spacecraft with the most fuel consumption. Therefore, a fuel-balanced problem is often mentioned in formation control. Spacecraft in the formation can be regarded as a kind of “virtual thruster” with three axis control and propose a balancing control algorithm based control allocation as to formation flying and each spacecraft in the formation. After that, we explain how to evaluate the performace of a balancing algorithm, such as fuel differences and lifetime of spacecraft formation, by defination of balance and sensitiveness indexes. Simulation results show that in cases of equal intial fuel and unequal intial fuel, formation reconfiguration control and thruster balancing control can improve the balance level of spacecrafts and thrusters, reduce sensitiveness under disturbations, and well robust to the variation of parameter.