Distributed Attitude Control And Vibration Suppression Of Flexible Spacecraft

Controlling through distributed actuator is becoming a focus in modern spacecraft technology with the upsizing of the spacecraft scale.The miniaturization and high performance of the developing actuator technology provides engineering foundation for distributed control.This thesis studies the attitude control and vibration suppression method of flexible spacecraft by scattered angular momentum exchange actuators.The main contents are as follows:For the scenario of using distributed actuators for vibration suppression,a decoupled design method for vibration suppression and attitude control is proposed.By constraining the generated torque of the actuator used for vibration suppression to be coplanar,there is no net torque output affecting the attitude control.The normal vector of the constraint plane is obtained according to the local angular velocity information of the actuator mounting node.The command torque of each actuator mounting node is gained by projection operation.The proposed method can be extended to the distributed application since it is easy to add or delete an actuator node in the coplanar operation.Modal analysis shows that the proposed method can increase the system damping and accelerate the vibration attenuation.For the scenario of using distributed actuators for simultaneous vibration suppression and attitude control,three distributed design methods are proposed.Firstly,the relationship of control allocation and vibration suppression is studied for the case that has a readymade attitude control law and uses scattered actuators for attitude control.In detail,the relationship of control allocation and system damping is studied.A vibration suppression method via control allocation is proposed.Thereafter,a local torque adjustment strategy in the sense of Lyapunov function is developed to further increase the system damping and guarantee the system stability.By formulating “increasing system damping” as the weight matrix in the objective function of an optimization problem,the above method can be realized in a distributed way.To reduce the dependence of the actuators on the central controller,by constraining each actuator mounting node applies an identical form of control law,an optimization method of vibration suppression and attitude control is proposed,which is based on the parameter allocation of the control law.The dynamic and steady-state performance of attitude control is realized by equality constraints on the parameters of control law.The system vibration is suppressed by minimizing an objective function associated with the vibration states of the actuator mounting nodes.The attitude control requirements can be achieved when the constraints of the optimization problem are satisfied.The optimal vibration suppression in the sense of the objective function is obtained when the optimization problem is solved.Subsequently,with the inspiration of the surplus based distributed optimization algorithm,a distributed realization of the proposed vibration suppression method via control parameters allocation is proposed.To further enlarge the optimization space,the distributed attitude control and vibration suppression under the game frame is studied.The vibration states at each actuator mounting node and the attitude of the spacecraft are taken to form the objective function of the game among the actuators.The results of the game are the optimal control policy that each actuator follows.The centralized solution of the game problem is given by statedependent state space model and Lyapunov iteration method.To enable each actuator mounting node to solve the game problem in a distributed way,the tracking differentiator is used to obtain the variables that have not been directly measured in the problem solving process.The value function approximation method based on neural network is adopted to solve the nonlinear differential equations in the game process.