| An indirect adaptive controller is developed for attitude control and momentum management of an evolutionary spacecraft. The proposed controller structure is applicable to a low-altitude, orbiting spacecraft that utilizes momentum exchange actuators for attitude stabilization. The control design, which is based upon the theory of nonlinear feedback linearization, represents the application of a (relatively) new theory of nonlinear control systems to a very real and complex problem facing spacecraft control engineers: namely, how to utilize momentum exchange actuators to stabilize a spacecraft while simultaneously minimizing stored momentum and continuously utilizing environmental gravity gradient torques to unload the momentum. The problem is solved here without making use of any classical linearization assumptions, such as negligible products of inertia or small angles and rates. An implicit function, whose roots define the singularities of the nonlinear control law, is determined, and its behavior around each equilibrium point (i.e., each torque equilibrium attitude) is examined. Since an evolutionary spacecraft is characterized by changing mass properties, a parameter identification scheme, which utilizes an extended Kalman filter, is added to the system for mass property estimation. The estimates of the inertia matrix are utilized to stabilize the spacecraft during changes in the mass properties. The combination of attitude control and momentum management is shown to be in direct conflict with mass property estimation and observability. Probing signals are added to the nonlinear control law to enhance the observability of the mass properties. A method for generating probing signals in the transformed states, based upon a desired response in the original coordinate system, is developed. Results show that the adaptive nonlinear controller stabilizes the International Space Station Alpha after Flight 40-12R during large mass property changes, successfully performs momentum management, and provides real time estimates of the mass properties in realistic high-fidelity simulations. |
