Observer-Based Control for High-Order Fully Actuated Systems

An observer-based control method for high-order fully actuated systems is proposed. First, a concept of exponentially stable observers is introduced, which, different from traditional state observers, requires not only asymptotic convergence of observation errors, but also exponential convergence. Further, inspired by existing results, two design methods for exponentially stable observers are developed, one of which is less conservative and the other is simpler and more straightforward to use. Secondly, a parametric control method based on the exponentially stable observer is proposed, which ensures the exponential stability of the closed-loop system. Moreover, the proposed method does not rely on the solution of nonlinear partial differential equations, and although the system is nonlinear and time-varying, the Separation Principle still holds under this control strategy, which is a significant advantage of the proposed method. Finally, the method is applied to the attitude control of flexible spacecraft with nonlinear inertia, and comparative simulation results verify the effect of the proposed approach.