| Our country's cruise missile technology has a big improvement in recent years, but there still exists a distance comparing with the international advanced level, especially the missile's guidance and control accuracy. In order to improve the accuracy of cruise missile attack, we must intensify the research on cruise missile's guidance and control technology. In this dissertation, cruise missile's guidance and control technology is studied based on bionic visual principle.First of all, the eagle's visual system, visual principle, visual field and et al. are studied. Based on the studied, a kind of cruise missile's image guidance method is studied based on bionic eagle visual principle. The method increases the observability of the guidance course.Then, eagle eyes lateral inhibition principle is studied. It is used for image edge extraction and image enhancement of cruise missile's image guidance. In order to improve the cruise missile's guidance accuracy, a kind of scene matching guidance method is studied based on the lateral inhibition principle's edge extraction. The method overcomes the image gray distortions or other aberrations'influence on cruise missile's guidance accuracy.Following, cruise missile's terminal guidance law is studied. A sliding mode variable structure guidance law for cruise missile based on neural networks is presented. The guidance law uses the self-learning ability of RBF neural networks and the robustness of variable structure control technology. It uses RBF neural networks to approximate the disturbance, and it overcomes the adverse influence of the unknown disturbance.Finally, cruise missile's flight control system is studied. Taking into account the existence of the cruise missile attitude angle layer's disturbance, a dynamic inversion flight controller based on RBF neural networks disturbance observer is designed for cruise missile. When taking into account the existence of the cruise missile attitude angle layer's disturbance and angular velocity layer's disturbance, a backstepping sliding mode flight controller based on RBF neural networks is designed for cruise missile. Simulation results show that the backstepping sliding mode flight controller based on RBF neural networks has a stronger anti-disturbance capability, and has a faster response speed. |