失效卫星远距离相对位姿估计与优化方法

牟金震; 刘宗明; 韩飞; 周彦; 李爽

doi:10.7527/S1000-6893.2021.24959

航空学报 >

2021 , Vol. 42 >Issue 11: 524959 - 524959

DOI: https://doi.org/10.7527/S1000-6893.2021.24959

论文

失效卫星远距离相对位姿估计与优化方法

牟金震 ,
刘宗明 ,
韩飞 ,
周彦 ,
李爽

展开

1. 上海航天控制技术研究所, 上海 201109;
2. 上海市空间智能控制技术重点实验室, 上海 201109;
3. 南京航空航天大学航天学院, 南京 211106;
4. 湘潭大学信息工程学院, 湘潭 411100

收稿日期: 2020-11-10

修回日期: 2020-12-03

网络出版日期: 2021-02-24

基金资助

国家重点研发计划（2016YFB0501003）；国家自然科学基金（61690214，11972182）；上海市科研专项（19511120900，19YF1420200）

收起

Long-range relative pose estimation and optimization of a failure satellite

MU Jinzhen ,
LIU Zongming ,
HAN Fei ,
ZHOU Yan ,
LI Shuang

Expand

1. Shanghai Aerospace Control Technology Institute, Shanghai 201109, China;
2. Shanghai Key Laboratory of Space Intelligent Control Technology, Shanghai 201109, China;
3. College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China;
4. College of Automation & Electronic Information, Xiangtan University, Xiangtan 411100, China

Received date: 2020-11-10

Revised date: 2020-12-03

Online published: 2021-02-24

Supported by

National Key Research and Development Program (2016YFB0501003); National Natural Science Foundation of China (61690214, 11972182); Shanghai Scientific Research Program (19511120900, 19YF1420200)

Fold

摘要

针对远距离非合作慢旋目标位姿估计精度问题，提出一种融合图像超分辨与视觉SLAM的相对位姿估计方法。算法主要包含3个步骤：通过梯度引导生成式对抗超分辨技术，提升目标图像的质量以获取更多更高质量的特征点；构建特征数据库实现当前帧与特征数据库的匹配，提升旋转目标的特征跟踪稳定性；利用图优化对多帧图像进行联合位姿优化，消除累计误差，得到更为精确的估计结果。为稳定网络的训练，将自然进化算法引入到对抗训练中。为增强模型的泛化性和鲁棒性，实验中的数据集采用半物理仿真获得。实验结果表明，当等效距离为25 m且失效卫星以25（°）/s的速度旋转时，目标图像经超分辨网络增强后，能够实现连续稳定的长时间测量。

关键词： 非合作目标; 失效慢旋目标; SLAM; 相对位姿估计; 生成式对抗网络; 图像超分辨

本文引用格式

牟金震 , 刘宗明 , 韩飞 , 周彦 , 李爽 . 失效卫星远距离相对位姿估计与优化方法[J]. 航空学报, 2021 , 42(11) : 524959 -524959 . DOI: 10.7527/S1000-6893.2021.24959

Abstract

To improve the accuracy of pose estimation of long-range non-cooperative targets with slow rotation, a method for relative pose estimation is proposed based on fusion of image super-resolution and visual Simultaneous Localization And Mapping (SLAM). The method mainly includes three steps. First, a gradient guidance Generative Adversarial Network (GAN)-based super-resolution model is utilized to improve the quality of images, so as to obtain more and higher quality feature points. Second, a feature database is constructed to match the current frame with the feature database, so as to improve tracking stability of the rotating target. Thirdly, pose graph optimization is carried out in multiple frames to optimize the joint pose, so as to eliminate the cumulative error and obtain more accurate estimation results. To stablize the training of GAN, an evolutionary algorithm is introduced. To enhance the generalization and robustness of the model, the dataset is obtained by semi-physical simulation. Experimental results show that when the imaging distance is equivalent to 25 m and the target is rotating at 25 (°)/s, our algorithm can realize continuous stable measurement after the images are enhanced by the super-resolution model.

Key words： non-cooperative target; failure rotating satellite; simultaneous localization and mapping (SLAM); relative pose estimation; generative adversarial network (GAN); image super-resolution

参考文献

[1] LI S, SHE Y. Recent advances in contact dynamics and post-capture control for combined spacecraft[J]. Progress in Aerospace Sciences, 2021, 120, 100678.
[2] 刘付成. 人工智能在航天器控制中的应用[J]. 飞控与探测, 2018, 1(1):16-25. LIU F C. Application of artificial intelligence in spacecraft[J]. Flight Control and Detection, 2018, 1(1):16-25(in Chinese).
[3] LI Y K, HAO X L, SHE Y C, et al. Constrained motion planning of free-float dual-arm space manipulator via deep reinforcement learning[J]. Aerospace Science and Technology, 2021, 109:106446.
[4] FLORES-ABAD A, MA O, PHAM K, et al. A review of space robotics technologies for on-orbit servicing[J]. Progress in Aerospace Sciences, 2014, 68:1-26
[5] 路勇, 刘晓光, 周宇, 等. 空间翻滚非合作目标消旋技术发展综述[J]. 航空学报, 2018, 39(1):021302. LU Y, LIU X G, ZHOU Y, et al. Review of detumbling technologies for active removal of uncooperative targets[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(1):021302(in Chinese).
[6] 周凡桂, 王晓光, 高忠信, 等. 双目视觉绳系支撑飞行器模型位姿动态测量[J]. 航空学报, 2019, 40(12):123059. ZHOU F G, WANG X G, GAO Z X, et al. Binocular vision-based measurement of dynamic motion for aircraft model suspended by wire system[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(12):123059(in Chinese).
[7] ZHANG J, DONG C C, ZHANG H, et al. Modeling and experimental validation of sawing based lander anchoring and sampling methods for asteroid exploration[J]. Advances in Space Research, 2018, 61(9):2426-2443.
[8] 刘宏, 刘冬雨, 蒋再男. 空间机械臂技术综述及展望[J]. 航空学报, 2021, 42(1):524971. LIU H, LIU D Y, JIANG Z N. Space manipulator technology:Review and prospect[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(1):524971(in Chinese).
[9] LIU L J, ZHAO G P, BO Y M. Point cloud based relative pose estimation of a satellite in close range[J]. Sensors, 2016, 16(6):824.
[10] DURRANT-WHYTE H, BAILEY T. Simultaneous localization and mapping:part I[J]. IEEE Robotics & Automation Magazine, 2006, 13(2):99-110.
[11] SCHNITZER F, JANSCHEK K, WILLICH G. Experimental results for image-based geometrical reconstruction for spacecraft Rendezvous navigation with unknown and uncooperative target spacecraft[C]//2012 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, NJ:IEEE Press, 2012:5040-5045.
[12] AUGENSTEIN S, ROCK S M, ENGE P, et al. Monocular pose and shape estimation of moving targets for autonomous rendezvous and docking[D]. Palo Alto:Stanford University, 2011.
[13] CHO D M, TSIOTRAS P, ZHANG G C, et al. Robust feature detection, acquisition and tracking for relative navigation in space with a known target[C]//AIAA Guidance, Navigation, and Control (GNC) Conference. Reston:AIAA, 2013.
[14] DOR M, TSIOTRAS P. ORB-SLAM applied to spacecraft non-cooperative rendezvous[C]//2018 Space Flight Mechanics Meeting. Reston:AIAA, 2018.
[15] TWEDDLE B E. Computer vision based navigation for spacecraft proximity operations[D]. Cambridge:Massachusetts Institute of Technology, 2013.
[16] THOMAS D, KELLY S, BLACK J. A monocular SLAM method for satellite proximity operations[C]//2016 American Control Conference (ACC). Piscataway:IEEE Press, 2016:4035-4040.
[17] 郝刚涛, 杜小平, 宋建军. 空间翻滚非合作目标相对位姿估计的视觉SLAM方法[J]. 宇航学报, 2015, 36(6):706-714. HAO G T, DU X P, SONG J J. Relative pose estimation of space tumbling non-cooperative target based on vision-only SLAM[J]. Journal of Astronautics, 2015, 36(6):706-714(in Chinese).
[18] 刘宗明, 曹姝清, 张宇, 等. 非合作航天器逆深度参数化姿态估计[J]. 光学精密工程, 2017, 25(2):451-459. LIU Z M, CAO S Q, ZHANG Y, et al. Inverse depth parametrization for attitude estimation of a non-cooperative spacecraft[J]. Optics and Precision Engineering, 2017, 25(2):451-459(in Chinese).
[19] 刘宗明, 牟金震, 张硕, 等. 空间失效慢旋卫星视觉特征跟踪与位姿测量[J]. 航空学报, 2021, 42(1):524163. LIU Z M, MU J Z, ZHANG S, et al. Visual feature tracking and pose measurement for slow rotating failure satellites[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(1):524163(in Chinese).
[20] 康国华, 马云, 乔思元, 等. 基于先验子图检测的失效航天器SLAM方法[J]. 中国空间科学技术, 2019, 39(1):1-10. KANG G H, MA Y, QIAO S Y, et al. SLAM method of failure spacecraft based on prior submap detecting[J]. Chinese Space Science and Technology, 2019, 39(1):1-10(in Chinese).
[21] 朱晏辰. 基于SLAM的非合作目标相对位姿测量研究[D]. 哈尔滨:哈尔滨工业大学, 2018. ZHU Y C. Research on measurement of relative pose for non-cooperative space targets based on SLAM[D]. Harbin:Harbin Institute of Technology, 2018(in Chinese).
[22] 周朋博, 刘晓峰, 蔡国平. 基于ORB-SLAM的低照度空间非合作目标的姿态估计[J]. 动力学与控制学报, 2021, 19(1):68-74. ZHOU P B, LIU X F, CAI G P. Attitude estimation of an non-cooperative spacecraft in low-light condition based on ORB-SLAM[J]. Journal of Dynamic and Control, 2021, 9(1):68-74(in Chinese).
[23] LEDIG C, THEIS L, F HUSZAR, et al. Photo-realistic single image super-resolution using a generative adversarial network[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017:105-114.
[24] WANG X T, YU K, WU S X, et al. ESRGAN:enhanced super-resolution generative adversarial networks[C]//European Conference on Computer Vision. 2018:63-79.
[25] MA C, RAO Y M, CHENG Y A, et al. Structure-preserving super resolution with gradient guidance[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway:IEEE Press, 2020:7766-7775.
[26] CAMPOS C, ELVIRA R, RODRÍGUEZ J J G, et al. ORB-SLAM3:an accurate open-source library for visual, visual-inertial, and multimap SLAM[J]. IEEE Transactions on Robotics, 2021:1-17.
[27] 周彦, 旷鸿章, 牟金震, 等. 面向半稠密三维重建的改进单目ORB-SLAM[J]. 计算机工程与应用, 2021, 57(8):180-184. ZHOU Y, KUANG H Z, MU J Z, et al. Improved monocular ORB-SLAM for semi-dense 3D reconstruction[J]. Computer Engineering and Applications, 2021, 57(8):180-184(in Chinese).
[28] HUI Z, LI J, GAO X B, et al. Progressive perception-oriented network for single image super-resolution[J]. Information Sciences, 2021, 546:769-786.
[29] 于浛, 魏喜庆, 宋申民, 等. 基于自适应容积卡尔曼滤波的非合作航天器相对运动估计[J]. 航空学报, 2014, 35(8):2251-2260. YU H, WEI X Q, SONG S M, et al. Relative motion estimation of non-cooperative spacecraft based on adaptive CKF[J]. Acta Aeronautica et Astronautica Sinica, 2014, 35(8):2251-2260(in Chinese).

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献