Observer-based L 2 -gain non-fragile guaranteed cost preview control for OSL-QIB nonlinear systems with input saturation

This paper develops an observer-based L 2 -gain non-fragile guaranteed cost preview controller (NFGCPC) for one-sided Lipschitz (OSL) and quadratically inner-bounded (QIB) nonlinear systems with input saturation. An augmented error system (AES) is constructed via state transformation to avoid direct differentiation of system states. By integrating transformed states, integral action, and previewed information, the AES converts the tracking problem into a saturated stabilization problem. A non-fragile controller is then synthesized to accommodate additive and multiplicative gain perturbations. Through Lyapunov stability analysis, linear matrix inequality (LMI)-based sufficient conditions are established to ensure asymptotic stability and L 2 -gain guaranteed cost performance. Observer and controller gains are determined concurrently. Furthermore, an optimization framework is formulated to minimize the cost upper bound and enlarge the estimated domain of attraction (DOA). Numerical examples demonstrate the effectiveness and superiority of the proposed approach.