Robust H_∞ output feedback control for Markovian jump systems with actuator saturation

Abstract

SUMMARYThis paper concerns the output feedback control problem for a class of uncertain continuous-time Markovian jump systems with actuator saturation. The controller is nonlinear in nature and will be first parameterized in the quasi-linear parameter varying form. Conditions under which the closed-loop system is stochastically stable with disturbance attenuation are then derived in terms of an LMI approach. The problem of designing an output feedback controller such that the estimate of the domain of attraction is enlarged is then formulated and solved as an optimization problem with LMI constraints. The case where the transition rate matrix of the Markov process is unknown is considered, and the robust H 1 output feedback controller is then derived. Finally, a numerical example is given to illustrate the effectiveness of the proposed results.

Positive observer design for discrete-time positive system with missing data in output

“…Though, α(k) is described by a Markov chain, it is said that system (2) or (3) is not a traditional Markovian jump system. In references [21,22,23,24,44,45,46,47], the system parameters switch synchronously according to a Markov process. Here, all the system parameters are deterministic, in which only the missing data in output is modeled into a Markov process.…”

“…Considering α ∈ [0, 1] and β ∈ [0, 1], it is known that (48) is guaranteed by (44) and (45), while (49) is implied by…”

Mode-dependent nonrational output feedback control for continuous-time semi-Markovian jump systems with time-varying delay

“…Note that most of the existing dynamic output feedback controller (DOFC) synthesis involves exclusively the rational controller for time-delay systems [28][29][30][31]. An attempt to design a nonrational controller [32], which is with a time-delay frame in the controller structure, seems more sensible and less conservative, since more information on time-delays is incorporated in the controller.…”

Finite-timeH∞synchronization control for semi-Markov jump delayed neural networks with randomly occurring uncertainties

“…On the other hand, jumping connection or switching topology often appears in a network owing to link failures or new creation [20,21,22,23,24,25,26,27]. Markov jumping parameters in studying synchronization of neural networks are wildly considered [28,29,31,30,32,33].…”