Periodically Time‐Varying Memory State‐Feedback for Robust H2 Control of Uncertain Discrete‐Time Linear Systems

This study is concerned with the synthesis of periodically time‐varying memory state‐feedback controllers (PTVMSFCs) for discrete‐time linear systems. In our preceding studies, we have already established a solid theoretical basis for linear matrix inequality (LMI)‐based (robust) H _∞‐PTVMSFCs synthesis, and the goal of this paper is to extend those results to the H ₂ performance criterion. In the H ₂ case, the main difficulty stems from the fact that we have to ensure the existence of common auxiliary variables for multiple LMI conditions that are related to the Lyapunov inequality and the inequalities for bounding traces that characterize the H ₂ norm. We can overcome this difficulty and derive a necessary and sufficient LMI condition for the optimal H ₂‐PTVMSFC synthesis. On the basis of this result, we also consider robust H ₂‐PTVMSFC synthesis for LTI systems with parametric uncertainties.     

Robust H 2 Filtering for Discrete‐Time Markovian Jump Linear Systems

This paper studies the problems of H₂ filtering for discrete‐time Markovian jump linear systems with non‐accessible mode information. The objective is to design a less‐conservative mode‐independent H₂ filter such that the filtering error system is stochastically stable. To this end, sufficient conditions for the existence of an upper bound of H₂ norm are given in terms of linear matrix inequalities. With the introduction of a slack variable, a less‐conservative filter is derived. The proposed robust H₂ filter design method is also applicable to cover the cases where the system matrices are subject to polytopic uncertainty, as well as the case where the transition probability matrix is subject to a given polytopic uncertainty. Application of the proposed method to three examples from the literature demonstrates the favorable performance of the proposed solution to existing approaches.     

Adaptive Observer‐Based Global Sliding Mode Control for Uncertain Discrete‐Time Nonlinear Systems with Time‐Delays and Input Nonlinearity

A new discrete‐time adaptive global sliding mode control (SMC) scheme combined with a state observer is proposed for the robust stabilization of uncertain nonlinear systems with mismatched time delays and input nonlinearity. A state observer is developed to estimate the unmeasured system states. By using Lyapunov stability theorem and linear matrix inequality (LMI), the condition for the existence of quasi‐sliding mode is derived and the stability of the overall closed‐loop system is guaranteed. Finally, simulation results are presented to demonstrate the validity of the proposed scheme.     

Dynamic Output‐Feedback Control for T‐S Fuzzy Systems with Input Time‐Varying Delay

This paper considers the problem of the control for T‐S fuzzy systems with input time‐varying delay via dynamic output feedback. Firstly, by applying the reciprocally convex approach, new delay‐dependent sufficient condition for performance analysis is obtained. Then, a less conservative condition for the existence of the controllers is given in terms of linear matrix inequalities (LMIs). Moreover, in the considered system, the time‐delay term is included in the measured output. This results in the difficulty in designing the controllers being increased and the obtained results being applied to a wider class of fuzzy systems than the most existing ones. The main contribution of this work lies in the application of the reciprocally convex inequality and the time‐delay term included in the measured output. Finally, the advantages and effectiveness of the present results are shown by several numerical examples.     

Practical adaptive fractional‐order nonsingular terminal sliding mode control for a cable‐driven manipulator

For the high precise tracking control purpose of a cable‐driven manipulator under lumped uncertainties, a novel adaptive fractional‐order nonsingular terminal sliding mode control scheme based on time delay estimation (TDE) is proposed and investigated in this paper. The proposed control scheme mainly has three elements, ie, a TDE element applied to properly compensate the lumped unknown dynamics of the system resulting in a fascinating model‐free feature; a fractional‐order nonsingular terminal sliding mode (FONTSM) surface element used to ensure high precision in the steady phase; and a combined reaching law with adaptive technique adopted to obtain fast convergence and high precision and chatter reduction under complex lumped disturbance. Stability of the closed‐loop control system is analyzed with the Lyapunov stability theory. Comparative simulations and experiments were performed to demonstrate the effectiveness of our proposed control scheme using 2‐DOF (degree of freedom) of a cable‐driven manipulator named Polaris‐I. Corresponding results show that our proposed method can ensure faster convergence, higher precision, and better robustness against complex lumped disturbance than the existing TDE‐based FONTSM and continuous FONTSM control schemes.     

Reliable sliding mode finite‐time control for discrete‐time singular Markovian jump systems with sensor fault and randomly occurring nonlinearities

This paper focuses on the problem of finite‐time $H_{\infty }$ control for one family of discrete‐time uncertain singular Markovian jump systems with sensor fault and randomly occurring nonlinearities through a sliding mode approach. The failure of sensor is described as a general and practical continuous fault model. Nonlinear disturbance satisfies the Lipschitz condition and occurs in a probabilistic way. Firstly, based on the state estimator, the discrete‐time close‐loop error system can be constructed and sufficient criteria are provided to guarantee the augment system is sliding mode finite‐time boundedness and sliding mode $H_{\infty }$ finite‐time boundedness. The sliding mode control law is synthesized to guarantee the reachability of the sliding surface in a short time interval, and the gain matrices of state feedback controller and state estimator are achieved by solving a feasibility problem in terms of linear matrix inequalities through a decoupling technique. Finally, numerical examples are given to illustrate the effectiveness of the proposed method.