Abstract
In this paper, a new design method which is based on the reduced-order algebraic Riccati equations is proposed to construct a near-optimal controller of multimodeling systems. First the existence of a unique and bounded solution of an algebraic Riccati equation is investigated. Furthermore, its asymptotic structure is also established. Second a new near-optimal controller is obtained which does not depend on the values of the small parameters. Then, it is proven that the near-optimal control can achieve a performance which is O(||μ||) close to the optimal performance. Finally, in order to evaluate the cost functional, we also propose a new algorithm which is based on the fixed point algorithm for solving the multimodeling generalized algebraic Lyapunov equation.
