In this paper, we apply a discrete-time Takagi-Sugeno Fuzzy Model (TSFM) based model predictive controller (MPC) to a Martian aerocapture vehicle following an arbitrary trajectory. We compare two baseline controllers: a continuous-time TSFM based parallel distributed controller (PDC) and a finite-horizon linear quadratic regulator (LQR). We evaluate the change in velocity (ΔV) required to bring the orbit of the controlled exit conditions to the orbit of the reference trajectory exit conditions over a range of initial condition errors and perturbations to atmospheric density. The LQR controller was least robust but performed best in a smaller range of perturbations. The PDC controller was most robust but performed the worst. The MPC based controllers demonstrate a balance of robustness and performance.

中文翻译：

---

基于Takagi-Sugeno模糊模型的再入车辆非线性模型预测控制

在本文中，我们将基于离散时间的Takagi-Sugeno模糊模型（TSFM）的模型预测控制器（MPC）应用于遵循任意轨迹的火星飞机。我们比较了两个基线控制器：一个基于连续时间TSFM的并行分布式控制器（PDC）和一个有限水平线性二次调节器（LQR）。我们评估的速度变化（Δ V）要求在一定范围的初始条件误差和对大气密度的扰动下将受控出口条件的轨道带入参考轨迹出口条件的轨道。LQR控制器的鲁棒性最差，但在较小的扰动范围内表现最佳。PDC控制器最坚固，但性能最差。基于MPC的控制器证明了鲁棒性和性能之间的平衡。