ABSTRACT This paper investigates a robust model predictive control (RMPC) scheme for linear parameter‐varying (LPV) systems with asymmetric actuator saturation and bounded disturbance. The saturation nonlinearity can deteriorate the control performance, and the asymmetric type is commonly encountered in engineering applications. To handle the asymmetry of saturation, a variable transformation method tailored to both positive and negative saturation levels of the actuator is carried out. The saturation range is fully considered to enlarge the searching space in theory to achieve less conservative results. Furthermore, an auxiliary control gain is employed to allow nonlinear saturation to take effect during operation. Based on the transformed model, an infinite‐horizon RMPC problem along with the corresponding optimization objective is designed. Cost upper bound constraint and invariant set condition are derived for the nonlinear saturated closed‐loop system. Less conservative results can be achieved from the conditions due to the full consideration and exploitation of the saturation range. We also confirm the recursive feasibility of the proposed algorithm and the closed‐loop stability of the saturated system. Eventually, two simulation examples are conducted to demonstrate the effectiveness of our RMPC scheme embedded with asymmetric saturation.
Wu et al. (Fri,) studied this question.