Switching Predictive Control for Continuous-time Markovian Jump Delay Systems

This paper is concerned with switching model predictive control (SMPC) for continuous-time Markovian jump delay systems (MJDSs). First, a piecewise constant switching predictive controller, which only depends on the average dwell time (ADT) switching laws rather than the jumping modes, is obtained by employing the ADT approach under the infinite-time predictive control design framework. Such a control strategy is proposed to make a trade-off between robustness and adaptivity when the design complexity of mode-independent and mode-dependent MPC is considered. It is revealed that the SMPC can deal with MJDSs with both time varying and time invariant jump rates and cover the mode-independent MPC as a special cease. Second, the feasibility of the SMPC scheme and the mean square stability of the closed-loop MJDS are discussed by using the stochastic invariance of the ellipsoid set over each sampling period. A numerical example is given to illustrate the main results.