Design of Gyro Stabilized Pod Control System
|School||Harbin Institute of Technology|
|Course||Control Theory and Control Engineering|
|Keywords||pod internal model control LMS adaptive Kalman filter|
Pod is an important Opto-Electronic tracking equipment used to track target. It’s widely used in national defence. For the bad time-delay influence to the system performance, which is caused by the image processing and transmission, the traditional design method can not satisfy the high precision need well when tracking maneuver target.In this dissertation, researches on the tracking loop controller designation and tracking error time-delay of the TV tracker pod are made. The principal works are summed as follows:(1) Based on broad references, the current research status is summarized.(2) The pod control system is designed as a whole. Then the motor, gyro and round inductosyn are selected. The transfer functions of TV tracker and control object in the tracking loop are established.(3) The tracking loop position controller is designed under the precondition that the tracking error time-delay is not considered. The position controller is designed with frequency lag correction and IMC correction. It’s proved that the maximal tracking error of the corrected system is less than 3 arc minutes when tracking equivalent sine signal.(4) Aiming to tracking error time-delay of TV tracker, LMS algorithm is used to predict and compensate it. Basic theory about LMS is discussed and its convergence condition is obtained. The simulation of azimuth orientation tracking loop prove that the maximal tracking error is less than 3 arc minutes with the algorithm.(5) Aiming to the disadvantages of LMS, adaptive Kalman prediction and filter algorithm is used in the designation. The target maneuver model in this dissertation is Current Statistical model. The tracking error time-delay is compensated by Kalman filter extrapolation. The simulation of azimuth orientation tracking loop prove that the maximal tracking error is less than 3 arc minutes with the algorithm.