Multi-view-based cooperative observation method for space non-cooperative target's attitude determination

基于多航天器协同观测的空间非合作目标姿态参数在轨识别方法研究

School of Astronautics, Northwestern Polytechnical University, Xi'an 710072, China

Received: 10 April 2020

Abstract

The parameter identification of a non-cooperative target is a key problem of on-orbit servicing. Due to poor priori information about the motion and structure of non-cooperative targets, the traditional cooperative estimation architecture is inadequate. In addition, the existing method based on single observation spacecraft is difficult to ensure a good observation state for a long time. This paper proposes a method for observing and estimating attitude kinematic state of a non-cooperative space target based on multi-spacecrafts cooperative observation. The space target's 3D point cloud data is reconstructed based on the data with different viewpoints. The target's attitude change is calculated based on the 3D point cloud matching, and the kinematics information is estimated via EKF filtering. The simulation results demonstrate the efficiency of the proposed approach. As the proposed method does not require priori model, structural information or accurate motion information of the tumbling non-cooperative target, it can avoid the influence of occlusion or date missing in space observation, it is suitable for space on-orbit servicing with non-cooperative target in future.

摘要

空间非合作目标参数识别是空间在轨服务的关键问题，由于非合作目标先验信息和合作标识的缺失，传统基于合作的空间目标参数识别方法难以适用；另外，现有的基于单一观测航天器的观测方法，难以长期保证良好的观测状态。提出了一种基于多航天器协同观测的非合作目标姿态运动学状态参数识别方法。采用多个航天器从不同视点位置观测非合作目标，基于多视点协作观测数据对目标进行三维重建，根据连续观测帧之间的目标点云匹配解算其姿态变化，最后通过EKF滤波算法求解出空间非合作目标的姿态参数。模拟仿真结果表明，给出的多航天器协同观测参数识别算法，适用于缺乏先验信息的非合作空间翻滚目标的姿态估计，同时避免了空间目标观测中遮蔽现象或特征跟踪丢失，可用于未来空间非合作目标在轨服务任务。