Scholarlink

<p>The Chang’E-6 (CE-6) mission’s first lunar farside samples return advances understanding of lunar evolution. This achievement highlights the necessity of rapid landing-trajectory reconstruction and precise landing-site characterization for mission execution. We present an intelligent vision-guided framework for lunar exploration, demonstrating that high-fidelity trajectory reconstruction can simultaneously enhance landing site localization accuracy and geological analysis precision. Our approach integrates: high-precision trajectory reconstruction from descent imagery, rapid localization utilizing deep learning-enhanced landmark matching achieving 0.90 m landing-site accuracy for Chang’E-6, and physically constrained geological analysis with visual rectification. By correcting camera distortions via pose parameters, we improve crater morphometry and surface age dating through orthorectified CSFD measurements. Our results reveal late-stage volcanism (&#xa0;~&#xa0;855 Ma) in Apollo crater ejecta, challenging early farside magmatic cessation models. Furthermore, we identify pristine anorthosite-pyroxene assemblages critical for probing lunar mantle differentiation. This work establishes an intelligent vision-guided approach to computational photogrammetry for extraterrestrial sample-return missions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent vision-guided trajectory reconstruction enables rapid localization and characterization of the Chang’E-6 landing site

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月17日

查看全文
  • Shihao Shu,
  • Liqun Lin,
  • Bingxu Hou,
  • Fangyi Cheng,
  • Dingquan Xue,
  • Chenfeng Hou,
  • Yong Lai,
  • Cai Meng,
  • Lu Li,
  • Yan Li,
  • Xiangzhi Bai

摘要

The Chang’E-6 (CE-6) mission’s first lunar farside samples return advances understanding of lunar evolution. This achievement highlights the necessity of rapid landing-trajectory reconstruction and precise landing-site characterization for mission execution. We present an intelligent vision-guided framework for lunar exploration, demonstrating that high-fidelity trajectory reconstruction can simultaneously enhance landing site localization accuracy and geological analysis precision. Our approach integrates: high-precision trajectory reconstruction from descent imagery, rapid localization utilizing deep learning-enhanced landmark matching achieving 0.90 m landing-site accuracy for Chang’E-6, and physically constrained geological analysis with visual rectification. By correcting camera distortions via pose parameters, we improve crater morphometry and surface age dating through orthorectified CSFD measurements. Our results reveal late-stage volcanism ( ~ 855 Ma) in Apollo crater ejecta, challenging early farside magmatic cessation models. Furthermore, we identify pristine anorthosite-pyroxene assemblages critical for probing lunar mantle differentiation. This work establishes an intelligent vision-guided approach to computational photogrammetry for extraterrestrial sample-return missions.