---

In this paper, a robust model predictive control (MPC) algorithm is designed for nonlinear uncertain systems in presence of the control input constraint. To achieve this goal, first, the additive and polytopic uncertainties are formulated in the nonlinear uncertain system. Then, the control policy is chosen as a state feedback control law in order to minimize a given cost function at each known sample-time. Finally, the robust MPC problem is transformed into another optimization problem subject to some linear matrix inequality (LMI) constraints. The controller gains are determined via the online solution of the proposed minimization problem in real-time. The suggested method is simulated for a second order nonlinear uncertain system. The closed-loop performance is compared to other control techniques. The simulation results show the effectiveness of the proposed algorithm compared to some existing control methods.
 

---

This paper is concerned with the problem of output feedback fuzzy mixed H2/H_inf tracking control design for nonlinear systems. A general and modified class of output-feedback controller structure is assumed. Using parallel distributed compensation, a fuzzy controller is proposed which not only satisfies an H_inf tracking control constraint, but also optimally minimizes a H2control performance measure. The proposed method leads to a trade-off between the tracking performance and the amount of control input effort. The problem formulation and the method of finding the optimal fuzzy tracking controller parameters involve a single step linear matrix inequalities form. On a benchmark example, the proposed method is applied on the inverted-pendulum system and is compared with traditional H_inf-only results from diffrent perspectives.
 

---

Abstract: This paper designs an optimal controller for the simultaneous determination of physical model parameters and LQR controller parameters. In some systems, it is possible to determine some of the model parameters by the designer. In conventional methods of optimal controller design for this group of systems, first, the model parameters are determined by the designer, and then in a separate step, the controller is designed for the definite model. In this paper, a method for the simultaneous determination of these two sets of parameters is presented for continuous-time linear systems. Simultaneous parameter determination is a nonlinear and non-convex optimization problem that in this paper a new method is considered to solve this problem. The non-convex optimization problem is transformed into a convex optimization problem by performing simplifications and then solved by the CVX toolbox of MATLAB software. The result is a controller with less control cost in comparison to conventional methods for this group of systems. By providing a simulation example, the performance improvement of the proposed method is shown.