Integrated guidance and control of the surface-to-air homing missile Pitch channel using optimal neural network

, the missile guidance and control system consists of three subsystems: navigation, guidance, and control. The task of these sub-systems is to calculate the deviation of the guided vehicle from the desired path so as to determine the appropriate movement or acceleration to compensate for the deviation. In the traditional methods , each of the guidance and control subsystems is designed separately, a. In the integrated guidance and control approach, the guidance law is developed separately and tested under the assumption of ideal autopilot. The autopilot is also designed independently and is tested under the assumption of an ideal guidance law. This paper describes the process of designing and simulating the performance of the optimal neural controller, which was created in order to guide the missile in a two-dimensional problem of minimizing the collision time and the distance from the target. In the design of the optimal neural controller, first the classical optimal neural controller (MLP) neural networks, the identifier, and the controller was designed and through simulation it was shown that the performance of this controller is not satisfactory. Therefore, by replacing the estimator MLP networks and controller with the deep type network, along with the use of the concepts of reinforcement learning, a quite improved performance was demonstrated through simulation. In this research, the integrated rocket model was made by integrating deep learning neural network with optimization algorithms, and the use of neural network control and optimization algorithms increased collision accuracy and reduced flight time.