Output Feedback Asynchronous Fuzzy SMC of Nonlinear Markov Jump Systems via Hidden Mode Detections
针对系统模式不可直接观测的非线性马尔可夫跳变系统，利用隐马尔可夫模型描述异步切换，提出一种基于观测模式的输出反馈动态滑模控制方法，保证系统稳定并满足耗散性能。
This work is concerned with the asynchronous output feedback sliding mode control (SMC) of stochastic nonlinear Markov jump systems (MJSs) via Takagi–Sugeno fuzzy models. Due to some real-world environment limitations, the actual system modes that are not directly available for controller synthesis are known as hidden modes. Then the sliding surface/sliding mode controller modes are featured as observable modes, and the relationship between these two concepts is established by employing emission probabilities. As a two-layer stochastic process, the hidden Markov model (HMM) governs the jump parameters and characterizes the asynchronous mode switching phenomenon between the original plant and the sliding surface/sliding mode controller. By integrating the sliding surface with the dynamical features of fuzzy MJSs, the dynamics of the sliding motion are described by constructing a T–S fuzzy singular MJS. Under a unified convexification setup, novel dissipative performance and stochastic stability analysis results on the sliding motion are proposed. In view of the full MJS states also not measurable, a novel observed-mode-based asynchronous output feedback dynamic SMC synthesis approach is propounded to ensure the MJSs’ states are located in a vicinity of the sliding surface. Illustrative simulation examples are finally provided to validate the superiority and effectiveness of the developed scheme.