Because the missile is autonomous, its control system must provide adequate flight stability while ensuring sufficient response to track commands. This paper is devoted to investigate the usefulness of the classical and modern control techniques for autopilot design for different evaluation approaches. The present work is concerned with improving the
performance of a surface-to-surface controlled aerodynamically guided missile system via both classical PID and predictive autopilots. The design and analysis necessitate somehow accurate system model with different uncertainties via 6-DOF simulation. The governing differential equations of the missile motion are derived with the aerodynamic model of the
missile constructed by means of the Missile Datcom software. After obtaining the required aerodynamic stability derivatives using the generated aerodynamic data, the necessary transfer functions are determined based on the equations of the missile motion. Next, the normal acceleration autopilot is designed using the determined transfer functions. The autopilot is designed to realize the command signals generated by the guidance laws which are in the
form of normal acceleration components. Using the entire system model, the computer simulations are carried out using the Matlab-Simulink software where the classical and predictive autopilots are compared via time response along the flight path