Masaya Tanemura

Linear Matrix Inequality Method for a Quadratic Performance Index Minimization Problem with a class of Bilinear Matrix Inequality Conditions

There are a lot of design problems of control system which are expressed as a performance index minimization under BMI conditions. However, a minimization problem expressed as LMIs can be easily solved because of the convex property of LMIs. Therefore, many researchers have been studying transforming a variety of control design problems into convex minimization problems expressed as LMIs. This paper proposes an LMI method for a quadratic performance index minimization problem with a class of BMI conditions. The minimization problem treated in this paper includes design problems of state-feedback gain for switched system and so on. The effectiveness of the proposed method is verified through a state-feedback gain design for switched systems and a numerical simulation using the designed feedback gains.

Parallel feedforward design using LMI for unknown disturbance estimation of non-minimum phase system

An unknown disturbance estimation method for non-minimum phase systems using a parallel feedforward model has been proposed. The parallel feedforward model is adopted in order to improve the estimation performance and it must ensure minimum phase property of the augmented system consisting of the plant and the parallel feedforward model. In addition, the frequency response of the parallel feedforward model must match to that of the plant. This paper derives design conditions of the parallel feedforward model as LMI conditions and designs the parallel feedforward model by solving a convex optimization problem.