宇航学报

宇航学报 ›› 2021, Vol. 42 ›› Issue (9): 1128-1138.doi: 10.3873/j.issn.1000-1328.2021.09.008

• 制导、导航、控制与电子 • 上一篇    下一篇

一种参数化非合作目标相对位姿和惯量估计方法

冯乾，潘泉，侯晓磊，杨家男   

  1. 1. 西北工业大学自动化学院，西安 710072；2. 西北工业大学信息融合教育部重点实验室，西安 710072
  • 收稿日期:2020-10-09 修回日期:2021-01-22 出版日期:2021-09-15 发布日期:2021-09-15
  • 基金资助:
    国家自然科学基金（61790552，62073264，61703343）；中央高校基本科研业务费（102018JCC003）

A Parameterized Method to Estimate Relative Pose and Inertia Information of a Non cooperative Target

FENG Qian, PAN Quan, HOU Xiao lei, YANG Jia nan   

  1. 1. School of Automation, Northwestern Polytechnical University, Xi’an 710072, China;2. Key Laboratory of Information Fusion Technology Ministry of Education, Northwestern Polytechnical University, Xi’an 710072, China

  • Received:2020-10-09 Revised:2021-01-22 Online:2021-09-15 Published:2021-09-15

PDF

291

摘要： 针对故障卫星、失效航天器、空间碎片等空间非合作目标无先验模型，且无法直接获得其角速度及惯量参数等问题，提出一种参数化非合作目标相对位姿和惯量参数估计方法。首先，基于自由航天器姿态动力学模型，用反双曲正切函数来描述含两个独立变量的惯量参数，建立非合作目标角速度传播模型方程；基于立体视觉测量模型，建立非合作目标上若干特征点的量测方程。然后，结合Clohessy Wiltshire方程描述的航天器间相对运动学模型，以非合作目标的相对位置、相对线速度、相对姿态、惯性角速度及惯量参数为状态量，设计扩展卡尔曼滤波器以估计各状态量。最后，进行不同场景的数值仿真验证。蒙特卡洛仿真结果表明，所设计的滤波器在不同噪声水平下能够高精度地有效估计出非合作目标的相对位姿和惯量参数。


关键词: 立体视觉, 非合作目标, 相对位姿, 惯量参数, 状态估计, 扩展卡尔曼滤波

Abstract: Aiming at solving the problem that space non cooperative targets, such as malfunctioning satellites, failed spacecraft and space debris, do not have a priori information for directly obtaining their models, inertia parameters and inertial angular velocities, a parameterized method based on stereo vision is proposed to estimate a non cooperative target’s pose and inertia information. Based on the equation of attitude dynamics for a free spacecraft, the inverse hyperbolic tangent functions with two independent variables are adopted to parameterize the inertia ratios of the non cooperative target, then the angular velocity propagation equation of the target is established. Measurement equations are established for the non cooperative target using several feature points acquired from the stereo vision measurement system. Whereafter, combined with the relative kinematics model described by Clohessy Wiltshire equation, an extended Kalman filter is designed to estimate the state, including the relative position, relative linear velocity, relative attitude, inertial angular velocity and inertia parameters of the non cooperative target. Finally, the performance of the proposed algorithm is investigated using numerical simulation in different scenarios. The Monte Carlo simulation results demonstrate that the designed filter can effectively estimate the relative pose and inertia parameters of the non cooperative target with high accuracy in various levels of measurement noise.


Key words: Stereo vision, Non cooperative targets, Relative pose, Inertia parameters, State estimation, Extended Kalman filter

中图分类号: