Skip to main navigation Skip to search Skip to main content

Pose measurement for non-cooperative target based on visual information

Research output: Contribution to journalArticlepeer-review

Abstract

In the process of on-orbit servicing, chasers achieve approach and contact with target spacecraft are via a series of orbital transfer and motion control. However, most target spacecraft are non-cooperative targets. As a matter of fact, mainly technical difficulty of non-cooperative targets still lies in unable directly obtaining their motion parameters, thereby, it is impossible to realize their autonomously rendezvous and capture. Thus, measure motion parameters of non-cooperative targets from outside is significant for research on-orbit service. In this paper, a pose measurement method coupled initial pose measurement with modelbased tracking is put forward for space non-cooperative targets. Specifically, an algorithm based on SIFT matching to measure initial pose of target tracking is proposed to start subsequent target tracking. Afterwards, a target tracking algorithm is presented, which is via a priori information and the combination of edge features and point features. As well as, an M estimator is introduced to improve the accuracy and robustness of target tracking algorithm. Finally, an experimental platform is constructed to fulfill experiments to validate the effectiveness of proposed method. Also, the experimental results demonstrate that our approach has highly precision and robustness in vary illuminations. Simultaneously, our method could satisfy the real-time requirement of pose measurement for non-cooperative targets.

Original languageEnglish
Article number2932835
Pages (from-to)106179-106194
Number of pages16
JournalIEEE Access
Volume7
DOIs
StatePublished - 2019

Keywords

  • Feature fusion
  • Model-based tracking
  • Non-cooperative target
  • Pose measurement

Fingerprint

Dive into the research topics of 'Pose measurement for non-cooperative target based on visual information'. Together they form a unique fingerprint.

Cite this