系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (8): 2802-2808.doi: 10.12305/j.issn.1001-506X.2026.08.25

• 制导、导航与控制 • 上一篇    

惯性信息辅助的星图检测识别技术

张前程1,2,3, 钟胜1,2, 李显成3, 李仲3   

  1. 1. 华中科技大学人工智能与自动化学院,湖北 武汉 430074
    2. 华中科技大学多谱信息处理技术国家级重点实验室,湖北 武汉 430074
    3. 华中光电技术研究所-武汉光电国家研究中心,湖北 武汉 430223
  • 收稿日期:2025-05-30 修回日期:2025-08-14 出版日期:2025-11-06 发布日期:2025-11-06
  • 通讯作者: 钟胜
  • 作者简介:张前程(1989—),男,高级工程师,博士研究生,主要研究方向为天文导航
    李显成(1993—),男,工程师,硕士,主要研究方向为星敏感器
    李 仲(1992—),男,高级工程师,博士,主要研究方向为天文导航

Inertial information-assisted star pattern detection and recognition technology

Qiancheng ZHANG1,2,3, Sheng ZHONG1,2, Xiancheng LI3, Zhong LI3   

  1. 1. School of Artificial Intelligence and Automation,Huazhong University of Science and Technology,Wuhan 430074,China
    2. National Key Laboratory of Science and Technology on Multi-Spectral Information Processing,Huazhong University of Science and Technology,Wuhan 430074,China
    3. Huazhong Institute of Electro-Optics-Wuhan National Laboratory for Optoelectronics,Wuhan 430223,China
  • Received:2025-05-30 Revised:2025-08-14 Online:2025-11-06 Published:2025-11-06
  • Contact: Sheng ZHONG

摘要:

针对空间强辐射、太阳与地气杂光强干扰和大气内颗粒物或大量发动机喷射物被太阳照亮等导致星敏感器测星图像存在大量假星目标,常规星图检测识别技术难以识别真星目标的问题,提出一种惯性信息辅助的星图检测识别技术。首先,利用惯性辅助信息预测星点在星敏感器靶面位置,依据导航星亮度和靶面预测位置将导航星分成多层并进行排序。然后,依次进行局部开窗采集疑似星点队列。最后,采用角距匹配和三角形极性匹配进行识别。试验证明,在假星目标数量高达真实星目标数量几十倍甚至数百倍条件下,真星目标依然能够正常检测识别,在高达4 000个假星目标条件下星图识别的成功率在98%以上,最大识别时间不超过1 s。该技术可有效解决大量假星条件下的星图检测识别问题。

关键词: 星敏感器, 惯性辅助, 星图检测, 星图识别, 假星目标

Abstract:

In view of the problem of a large number of false star targets in star tracker images caused by strong space radiation, strong interference from solar and earth-atmosphere stray light, and particles in the atmosphere or a large amount of engine spray being illuminated by the sun, conventional star pattern detection and recognition technologies are difficult to identify real star targets, a star pattern detection and recognition technology based on inertial information assistance is proposed. Firstly, inertial assistance information is used to predict the position of stars on the target surface of the star tracker, and the navigation stars are divided into multiple layers and sorted based on the brightness and the predicted position of the navigation stars. Then, partial windows are sequentially opened to collect suspected stars. Finally, angle distance and triangle polarity matching are used for recognition. Experiments show that even when the number of false star targets is tens or even hundreds of times higher than that of real star targets, the real star targets can still be detected and recognized. Under the condition of up to 4000 false star targets, the success rate of star map recognition is over 98%, and the maximum recognition time does not exceed 1 s. This technology can effectively solve the problem of star pattern detection and recognition under the condition of massive false stars.

Key words: star tracker, inertial assistance, star pattern detection, star pattern recognition, false star target

中图分类号: