robust model predictive control design using uncertainty estimation

This paper presents a robust model predictive control for the design of autopilots, specifically targeting unstable systems. these systems are highly sensitive to model uncertainties, necessitating an increase in control gain to effectively manage this sensitivity. however, such increases can jeopardize the stability margin and potentially induce system instability. to address this challenge, parameter estimation techniques are employed to enhance the robustness of the control strategy.the proposed approach consists of two main steps: first, the uncertain estimation of the input matrix, and second, the optimization over a finite horizon based on the estimated input matrix. an uncertainty matrix is derived using the extended kalman filter method and incorporated into the design of the predictive controller. this approach greatly enhances the system's robustness to uncertainties. furthermore, to handle the computational complexity associated with implementing predictive control, particularly for fast systems, the expansion of laguerre functions is utilized. a comparative analysis is conducted to evaluate the effectiveness of this approach against traditional pid and generalized incremental predictive control (gipc) methods, demonstrating its superior performance across various scenarios.