In this study, a quasi-continuous high-order sliding mode control approach is presented for the longitudinal model of a generic hypersonic flight vehicle with parametric uncertainties and actuator faults. The quasi-continuous high-order sliding mode controller is designed to track the responses of the normal system to guarantee the velocity and altitude track the reference trajectories rapidly. An improved measure by increasing the relative degree of the quasi-continuous high-order sliding mode is introduced to eliminate the effects on outputs caused by chattering and parametric uncertainties. In view of actuator coupling, an equivalent canonical model is formulated through feedback linearization to accelerate the faults estimation for the actuator faults system. A neural network observer is then utilized to online estimate the unknown faults. This observer can be applied to highly nonlinear system without any prior knowledge of system dynamics as it uses a nonlinear-in-parameters neural network. Meanwhile, the stability and convergence of the faults system is proved theoretically. Simulation results are presented to testify the effectiveness and robustness of the proposed control scheme.
HIGH ORDER SLIDING MODE CONTROL WITH ANTI-SWAY BASED COMPENSATION ON ARTIFICIAL NEURAL NETWORK BY PSO ALGORITHM FOR OVERHEAD CRANE
