The sliding mode disturbance compensated GPC method for a class of uncertain systems
一类不确定系统基于滑模干扰补偿的广义预测控制

The generalized predictive control(GPC) method is an optimal control strategy which is based on the mechanism model. However, if the system has internal uncertainty, model error and external disturbance, the performance of the system which is controlled by the GPC method based on the nominal model will be degraded. Therefore, to deal with the uncertain nonlinear systems, we first design the ideal GPC law based on uncertain model, and then, we develop a slidingmode disturbance compensator(SMDC) to estimate the hybrid disturbance, its outputs are integrated with the nominal GPC law to compensate the uncertainty. The Lyapunov theory is used to analyze the stability of the compound closed-loop system. This strategy was applied to the design of a hypersonic vehicle attitude control system, and the simulation result indicates an excellent robust performance and desirable disturbance attenuation.