乱気流の事前情報を用いたロバストモデル予測制御による Gust Alleviation 制御

Abstract

This paper addresses the design problem of Gust Alleviation (GA) flight controllers for linearized longitudinal aircraft motions with some uncertainties using prior turbulence information via Model Predictive Control (MPC) scheme. Considering that the plant uncertainties are assumed to be modeled as time-invariant uncertain but bounded delays at the plant control input, we derive a plant set, the number of whose elements are finite, to represent the uncertainties without introducing any approximations. For this set, we derive a new formulation to obtain an optimal control input, which guarantees some robust performance with respect to GA performance against the delays, as a Second-Order Cone Programming (SOCP) problem. As the conditions in SOCP problems have the convexity with respect to the decision variables, the global optimal control input for our addressed problem is obtained using some effective software. Exploiting that our proposed method introduces no approximations when deriving the plant set and SOCP problems can give the global optima, we propose a method to identify whether or not the prior gust information improves GA performance. A numerical example which illustrates our conclusions is included.