Study On Large Angle Maneuver Of Multi-body Spacecraft

The large angle maneuver is one of the key technologies for spacecraft attitude control, which has been the focus of attention. The paper systematically studies large angle maneuver control of spacecraft in the presence of mass parameter uncertaintiesJourdain's principle is employed to derive the dynamic equations of multi-body spacecraft with flexible appendages, and the rotation angle of the end of antenna arm caused by the deformations is considered, which effects the tracking accuracy of the antenna. Introducing an angular velocity constraint and an angular constraint, two methods are proposed to compute the reference attitude profiles of the camera and antenna, respectively.To simplify the control design problem, this paper derives the desired inverse system, which can convert the attitude tracking problem into the regulator problem. Based on the desired inverse system and sliding mode control, a robust attitude tracking controller is developed in the presence of mass parameter uncertainties and external disturbance. By Lyapunov stablity theory, the closed loop system stability can be achieved. Base on the inverse system method and H_∞control method,μsynthesis technique, the paper designs two nonlinear robust controllers. The controller consists of a linear robust controller and a PD controller. The PD controller avoids the singular problem in the design of Mixed Sensitivity Robust Control.Based on radial basis function neural network and echo state network, two neural controllers are developed, respectively. The RBF controller consists of a radial basis function network and a robust controller. The radial basis function network can compensate the parameter uncertainties and external disturbances, and a robust controller can compensate RBFN approximation error and realize the anticipative stability and performance properties. Utilizing the finite time convergent property of terminal sliding mode, an online learning algorithm based on TSM for updating all the parameters of the RBFN is derived, which improves tracking performance of the controller. Simulation results demonstrate the good tracking performance of the control scheme. The robust controller based on echo state network is developed for the attitude tracking. The genetic algorithm is employed to optimize the ESN's primary parameters, which removes the difficulty of choosing the ESN parameters. This control approach requires no prior knowledge about the dynamic model.The paper designs a physical simulation experiment system for the rigid-flexible-fluid coupling multi-body spacecraft. Using law of equilibrium in connected vessels, the liquid fuel tank system is designed. The couple of rapid maneuver antennas are drived by planet gear synchronous driving apparatus, which are antisymmetric about the axis of magnetic-bear simulator. The design avoids the eccentric load, and ensures that the magnetic-bear simulator can work well. By choosing controllers, sensors and exciters, many experiment strategies can be realized, including the dynamic characteristics of the rigid-flexible-fluid coupling multi-body spacecraft, the large angle maneuver of the center body of spacecraft and multi-target attitude tracking control, and so on.