Constrained missile autopilot design based on model predictive control

摘要

A new controller is proposed for a type of typical nonlinear missile autopilots using model predictive control method in the presence of constraints. Nonlinear model is first transformed into a linear structure, i.e. the form of state-dependent coefficient, which is used as the internal model for prediction. Then the constrained solution is obtained by solving an online quadratic programming problem at each sampling time, hence practical performances can be guaranteed. The resulting control law ensures nominal acceleration tracking for the missile. The closed-loop system has a good robustness against disturbances. Compared to the proportional integral controller, the proposed controller is more suitable to implement in practice. Simulation results confirm the effectiveness of the proposed control strategy.