论文部分内容阅读
为解决机载光电跟踪系统在跟踪过程中由于工作平台框架的转动导致成像画面的旋转问题,提出了一种基于自适应神经网络的消旋控制方法。系统以消旋指令角作为给定位置信息,以光电编码器实测角度值前后两拍之差作为实测速度值,组成速度反馈内环;以陀螺仪测得的角度值作为位置反馈值,构成位置外环;校正算法采用二阶超前-滞后校正并加入了自适应神经网络算法对其控制参数进行自适应调整。实验结果表明,在消旋拍摄过程中,消旋速度满足设计要求,拍摄图片清晰,消旋精度(均方值)达到1.4’,比传统校正方法输出误差减少了46%。
In order to solve the problem of rotation of the imaging frame caused by the rotation of the working platform frame in the tracking process of the airborne electro-optical tracking system, an adaptive racemization control method based on neural network is proposed. The system uses the anti-rotation command angle as the given position information, and takes the difference between the two measured before and after the angle value measured by the optical encoder as the measured speed value to form the speed feedback inner ring; the angle value measured by the gyroscope is used as the position feedback value to form the position Outer loop; the correction algorithm uses second-order lead-lag correction and adds adaptive neural network algorithm to adjust its control parameters adaptively. The experimental results show that the racemization speed meets the design requirements in the meridional shooting process. The captured images are clear and the racemization accuracy (mean square value) reaches 1.4 ’, which is 46% less than that of the traditional calibration method.