Predictive Functional Control-Based Missile Autopilot Design

Abstract

Predictive functional control is applied to design a missile autopilot considering control constraints. The missile nonlinear dynamics are first transformed into a linear structure with state-dependent coefficient matrices. At each sampling instant, the internal state-space model for prediction is obtained through a normal discretization procedure. Based upon this model, a predictive functional controller for the nonlinear missile autopilot is proposed. Compared with the conventional predictive control algorithms, which usually involve an online quadratic programming in the practical implementation, the new controller demands less online computation resources. Simulation results validate the effectiveness and robustness of the proposed algorithm.