一种临近空间飞行器动态拖尾星图复原方法

A dynamic trailing star map restoration method of near space vehicle

临近空间飞行器飞行速度快,机动性强,飞行过程中的大姿态机动会导致星敏感器拍摄的星图出现拖尾现象,从而使天文导航的计算结果出现较大的误差甚至失效。针对这一问题,采用IMU量测信息辅助设计动态拖尾星图复原算法,针对传统维纳滤波无法对光轴向角速度引起的拖尾进行有效复原的缺陷,提出了一种图像分割分布式拖尾星图复原方法。通过图像分割对不同图像区域的星点进行分布式维纳滤波,完全适应三轴角速度变化。搭建了天文导航虚拟验证仿真平台并进行了仿真验证。仿真结果表明,图像分割分布式拖尾星图复原方法可对拖尾星图进行有效复原,复原后姿态精度可维持在20″(3σ)以内,质心提取精度提高20%,姿态精度提高10%。所提出的方法能有效提高动态环境下天文导航系统的适应能力,对未来临近空间飞行器天文导航算法的设计具有一定的参考价值。

The near space vehicle has fast flight speed and strong maneuverability,and the large attitude maneuver will lead to the tailing of star map,which will increase the error of celestial navigation and even cause the algorithm invalid.A dynamic trailing star image restoration method base on the measurement of IMU is proposed to solve the problem that traditional Wiener filter can not effectively recover the tail caused by the optical axial angular velocity.In the proposed method,star points in different image regions are restored by distributed Wiener filter,which can fully adapt to the change of triaxial angular velocity.The simulation platform of celestial navigation virtual verification is built,and the simulation results show that the image segmentation distributed trailing star image restoration method can effectively restore the trailing star image,the attitude accuracy can be maintained within 20"(3σ)after restoration,the accuracy of centroid extraction is increased by 20%,and the attitude accuracy is increased by 10%.The proposed method can effectively improve the adaptability of celestial navigation system in dynamic environment,and provides a reference for the design of near space vehicle celestial navigation in the future.