Study Of Tracking Control And Miss Distance Prediction For Electro-optical Stabilization Image System

In modern war, the airborne weapon system not only requires high precision and long distance fire range, but also can work under the rough conditions no matter day or night. The swinging of the carrier is isolated by the electro-optical stabilized system, which is stable relative to the inertial space. And it will realize searching, capture, tracking and orientating functions to the object under the control command. With the increase of the velocity and acceleration of the moving target, it produces a large increase of the electro-optical stabilization image system. How to improve the tracking precision has become a research highlight recently.Firstly, developments of tracking control and miss distance prediction methods of electro-optical image stabilized system have been summarized and significances of the research have been notified in the dissertation. Then, mathematic models of the servo system are built up and simulation diagrams are established with Simulink. These models include two axes servo systems: speed loop and position loop. It provides the basis of study for designing the servo systems. The key factor of improving the ability of electro-optical image stabilized system to tracking ability is to reduce setting time and overshoot. Classic control was used to improve their dynamic performances. Aiming at the difficulty of pure long time delays of miss distance, we can get result that the longer time delay, the larger tracking error are produced. In order to overcome the bad effect of long delays, several prediction filter methods are introduced in the study. All of these methods can diminish the dynamic error greatly. Kalman filter for prediction is the most applicable method by comparing with simulation process and results.At the end of the paper, application of Kalman filter method in the electro-optical stabilized system is discussed in detail. The reason that miss distance prediction method can improve tracking precision is analyzed theoretically. Effects on dynamic performances of the method are analyzed. The simulation of different filters have been done by Matlab.