Adaptive nonlinear control for a reusable space vehicle’s longitudinal motion under model uncertainties and physical constraints

Abstract

The main object of this thesis is the Lyapunov-based adaptive controller design for the nonlinear longitudinal dynamics of an air-breathing Reusable Space Vehicle (RSV) under model uncertainties and physical constraints. The RSV is highly unstable by model uncertainty and external disturbance. Model uncertainty includes the variation of aerodynamic coefficients, for which high fidelity modeling is not feasible in a hypersonic region due to the technical limitation of the ground experimental analysis. Additionally, the external disturbances such as wind gust, longitudinal moment, force perturbations and thrust misalignment should be considered in the control system of the hypersonic flight. In general, it is very difficult to precisely identify all nonlinearities, and the complex nonlinear aerodynamic characteristics of a RSV. RSVs usually experience rapid variation of flight conditions. In such a flight envelope, RSVs are subjected to abnormal flight conditions such as aerodynamic parameter uncertainty, force and moment perturbation, inaccurate states measurements at hypersonic velocity, saturation of critical control components and so on. Such flight conditions are regarded as sensitive elements to operation, causing adverse effects on the performance of flight control. First, the governing equations of motion of RSV’s longitudinal dynamics are derived and the longitudinal dynamics including thrust misalignment also are proposed. The misalignment terms are considered as disturbances during numerical simulation. Then, a Lyapunov based adaptive control by using a constrained command filtered and tuning functioned approach is proposed. A minimized online parametric adaptation law using a tuning function is designed to compensate for the tracking errors by the variation of model uncertainties and external disturbance. A first-order constrained command filter is designed to handle the physical constraints of command signals and estimate the partial derivatives of ...