Adaptive self-Triggered model predictive control of discrete-Time linear systems

The communication load is a main concern for implementing model predictive control (MPC) to networked control systems. This paper investigates the self-Triggered MPC problem for discrete-Time linear systems under network environments to reduce communication load. We first propose a novel cost function which explicitly takes the communication cost into consideration. Based on the new cost function, an adaptive self-Triggered MPC strategy is designed in the sense that the self-Triggered time intervals can be adaptively adjusted via optimization. We further develop a simplified approach to solving the optimal control sequences and the optimal self-Triggered time intervals. The stability of the closed-loop system is analyzed and we show that the closed-loop system is asymptotically stable if some mild controllability conditions are satisfied. Finally, the effectiveness of the designed algorithm is verified via simulation studies.