A novel two-loop large offset tracking control of an uncertain nonlinear system with input constraints

Abstract Under a large deviation between the initial state and the desired trajectory, the traditional single-loop control method usually fails to solve the output tracking of SISO nonlinear systems due to the inherent uncertainties, input saturation, and unmeasurable states. In this work, a novel tracking control system is proposed in the framework of two-loop control system to achieve high-accurate tracking performance and reconcile the contradiction between the system's available regulating capacity and the required tracking capacity. Specifically, an observer and command-filtered-based adaptive fuzzy controller are synthesized in the inner loop to guarantee a satisfying tracking accuracy and remain robust to uncertainties. A simplified fuzzy approximator is utilized to estimate the uncertainties with a single adaptive parameter. Then, a replanned reference trajectory is generated in the outer loop by online optimization to minimize the transient performance over the predictive horizon. Moreover, a tracking governor of reference trajectory is proposed to obtain an adaptive convergence rate with respect to the desired trajectory while only satisfying relatively loose constraints. Simulation results prove that the proposed two-loop scheme can not only improve the tracking accuracy; but also has the adaptability to fulfill the particular engineering problems that have a large difference between the initial condition and the desired trajectory.