為不確定非線性系統設計強健模糊控制器，本文提出一個綜合的設計方法。首先，將不確定非線性系統，模型化成高木－菅野模糊模型。其次，結合強健穩定性理論，T-S模糊模型技術，和模型預測控制策略，在每一個取樣時間，透過極小化模型預測控制之無限水平目標函數的最小上界，可計算出預測狀態回授控制器增益。此外也導出滿足輸入限制條件，具有強健漸近穩定之充分條件，並化成線性矩陣不等式。最後，對不穩定又非線性且具有時變參數之卡車拖車系統，作倒車控制試驗。從數值模擬結果證明，所提出的方法有效，並可達成強健模糊模型預測控制的目標。
This paper proposes a synthetic scheme for the robust fuzzy predictive controller design of uncertain nonlinear systems. First, the Takagi-Sugeno (T-S) fuzzy model is employed to modeling the uncertain nonlinear systems. Next, combining robust stability concept, T-S fuzzy model approach, and model predictive control (MPC) strategy, at each sampling time, the predictive state feedback controller is calculated by minimizing a least upper bound for MPC infinite horizon objective function. Sufficient conditions, satisfying the input constraints, are also derived for robust asymptotic stability and can be reduced into linear matrix inequalities (LMIs). Finally, a truck-trailer system which has unstable and nonlinear with time varying parameter is applied for backing-up control test. The simulation result demonstrates that the proposed approach can achieve the robust fuzzy MPC objective.
