Study on Enlarging the Searching Scope of Staring Area and Tracking Imaging of Dynamic Targets by Optical Satellites

Conventional optical satellite imaging, such as push-broom imaging, obtained static image and was unable to get the characteristic of dynamic target. Staring imaging can obtain dynamic image. However, the coverage of the image, limited by the field of view (FOV) of the payload, is within only a dozen square kilometers. In this paper, a regional searching model and a tracking algorithm are proposed to enlarge the searching scope of staring area and to instruct tracking imaging of dynamic targets with a small FOV. In order to enlarge the searching area, a mathematical model is designed by setting a virtual target position under the original target position. Considering the center of dynamic target as the target-missing quantity, the tracking algorithm is established to control the satellite attitude so as to ensure that the optical axis points to the target-missing quantity. Finally, the algorithm is verified through ground experiments and the conclusion is demonstrated through on-orbit experiments with the first video satellite of Jilin province(JL-1). Compared with conventional staring imaging, the range of searching imaging can be expanded by 8 times. When the attitude angle and angular velocity of the satellite are better than 0.05° and 0.08°/s, respectively, the tracking accuracy of the target center can be shorter than one pixel, and the image will be distinctly visible.