摘要
为加速失效卫星在轨操作实验平台中的图像处理,提出一种基于运动辅助的视觉导航处理方法。首先,根据在轨操作实验平台的工作原理,建立了视觉导航系统与失效卫星模拟装置之间的相对运动模型;然后,利用扩展卡尔曼滤波器估计两者之间的相对位置与目标运动角速度;最后,根据运动模型与坐标系之间的转换关系,预测失效卫星模拟装置特征在视觉系统中的像素坐标,在特征临近区域进行局部图像处理,可提高视觉图像处理速度。实验结果表明:采用文中方法较改进前的图像处理速度提高了3倍,从而验证了运动辅助的视觉导航处理方法的有效性。
For speeding up the image processing of failure satellite on orbit operation experiment platform, a visual navigation processing method which is based on motion aided is proposed. Firstly, on the basis of the orbit operation experiment platform, the relative motion model between the visual navigation system and the failure satellite simulator is established. Secondly, use extended Kalman filter to estimate the relative position and the target angular velocity. Lastly, according to the transformation relationship between the motion model and the coordinate system, predict the pixel coordinates of the characteristic of the failure satellite simulation device in the vision system. Then perform the local image processing in the feature proximity area. And it can speed up the image processing. The experiments result shows that the proposed method can improve the image processing speed by 3 times. And the effectiveness of the visual navigation processing method has been verified by the experiment.
出处
《兵工自动化》
2017年第10期34-37,共4页
Ordnance Industry Automation
基金
国家自然科学基金(61203197)
关键词
失效卫星
运动辅助
图像处理
failed satellites
motion aided
image processing
