基于模型定义的光学遥感卫星星上处理系统设计

doi:10.11947/j.AGCS.2024.20220658

摘要/Abstract

摘要：

本文提出了一种基于模型定义的光学遥感卫星星上处理系统设计方法，构建了“硬件资源-算子模块-通用处理核-典型应用”的星上处理任务行为模型范式。在硬件层面，采用星上异构嵌入式计算平台，将多处理器一体化设计，通过标准化高速数据互连，完成高速信号传输和数据处理；通过网络拓扑实现良好的扩展性，支持设备规模及数据处理复杂性的变化；在软件层面，针对星上智能处理任务中频繁数据读写操作、大量重复计算操作、卷积神经网络的通用计算与加速等需求，设计了星上平台指令集、星上算法通用算子和星上智能网络组件组成的可配置算子模块，并基于该模块可快速实现特定算法的硬件IP核。经仿真试验验证，本文方法可根据卫星平台和星上处理任务需求，按需适配最佳软硬件解决方案，并有效提高了计算资源利用率；提出的结合云检测的实时流水压缩编码方案，显著提升了压缩性能；设计的轻量化目标检测识别方法，计算资源效率达到91.5%；以高分一号、高分七号原始数据率为例进行分析，相比常规方案整体计算资源利用率分别提高了16.51%、17.77%。实现了星上处理系统研制过程模型化、工作模式可定义、逻辑资源共享化，是适应卫星小型化、快速部署的一种合理优化选择。

关键词: 遥感卫星, 星上处理, 模型定义, 按需适配, 异构计算

Abstract:

In this paper, a design method of optical remote sensing satellite on-board processing system based on model definition and model-based systems engineering (MBSE) is proposed, and a model paradigm of on-board processing task behavior of “hardware resource-operator module-general processing core-typical application” is constructed. At the hardware level, the on-board heterogeneous embedded computing platform is adopted to integrate the design of multiple processors, and complete high-speed signal transmission and data processing through standardized high-speed data interconnection. In order to support changes in device scale and data processing complexity, good scalability through network topology should be achieved. At the software level, in view of the requirements of frequent data read and write operations, a large number of repeated computing operations, and general computing and acceleration of convolutional neural networks in on-board intelligent processing tasks, a configurable operator module composed of the instruction set of the on-board platform, the general operator of the on-board algorithm and the components of the on-board intelligent network is designed, and the hardware IP core of the specific algorithm can be quickly implemented based on the module. Simulation experiments show that this method can adapt the best software and hardware solutions according to the needs of satellite platforms and on-board processing tasks, and effectively improve the utilization rate of computing resources. The proposed real-time streaming compression coding scheme combined with cloud detection significantly improves the compression performance. The designed lightweight target detection and recognition method achieves a computing resource efficiency of 91.5%. Taking the raw data rates of GF-1 and GF-7 as examples, the overall computing resource utilization rate increased by 16.51% and 17.77% respectively compared with the conventional solution. The method has realized the modeling of the development process, the definable working mode, and the sharing of logical resources of on-board processing system, which is a reasonable optimization choice to adapt to the miniaturization and rapid deployment of satellites.

Key words: remote sensing satellite, on-board processing, model-defined, on demand adaptation, heterogeneous computing

中图分类号:

P237

刘薇, 刘松林, 郭子博, 刘凯, 张荔哲. 基于模型定义的光学遥感卫星星上处理系统设计[J]. 测绘学报, 2024, 53(4): 689-699.

Wei LIU, Songlin LIU, Zibo GUO, Kai LIU, Lizhe ZHANG. Design of optical remote sensing satellite onboard processing system based on model definition[J]. Acta Geodaetica et Cartographica Sinica, 2024, 53(4): 689-699.

图/表 14

图1

图2

表1

图3

图4

图5

图6

图7

表2

图8

图9

图10

表3

表4

参考文献 34

[1]	杨元喜, 王建荣, 楼良盛, 等. 航天测绘发展现状与展望[J]. 中国空间科学技术, 2022, 42(3):1-9.
	YANG Yuanxi, WANG Jianrong, LOU Liangsheng, et al. Development status and prospect of satellite-based surveying[J]. Chinese Space Science and Technology, 2022, 42(3):1-9.
[2]	杨元喜, 王建荣. 泛在感知与航天测绘[J]. 测绘学报, 2023, 52(1):1-7.DOI: 10.11947/j.AGCS.2023.20220405.
	YANG Yuanxi, WANG Jianrong. Ubiquitous perception and space mapping[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(1):1-7.DOI: 10.11947/j.AGCS.2023.20220405.
[3]	王建荣, 杨元喜, 胡燕, 等. 光学测绘卫星现状与发展趋势分析[J]. 武汉大学学报（信息科学版）, 2023, 48(3):333-338.
	WANG Jianrong, YANG Yuanxi, HU Yan, et al. Analysis on status quo and development trend of optical surveying and mapping satellites[J]. Geomatics and Information Science of Wuhan University, 2023, 48(3):333-338.
[4]	王密, 仵倩玉. 面向星群的遥感影像智能服务关键问题[J]. 测绘学报, 2022, 51(6):1008-1016.DOI: 10.11947/j.AGCS.2022.20220186.
	WANG Mi, WU Qianyu. Key problems of remote sensing images intelligent service for constellation[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(6):1008-1016.DOI: 10.11947/j.AGCS.2022.20220186.
[5]	杨元喜, 任夏, 王建荣. 集成型与智能型测绘卫星工程发展及其关键技术[J]. 测绘学报, 2022, 51(6):854-861.DOI: 10.11947/j.AGCS.2022.20220048.
	YANG Yuanxi, REN Xia, WANG Jianrong. Development of integrated and intelligent surveying and mapping satellite project with corresponding key technology[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(6):854-861.DOI: 10.11947/j.AGCS.2022.20220048.
[6]	江碧涛. 我国空间对地观测技术的发展与展望[J]. 测绘学报, 2022, 51(7):1153-1159.DOI: 10.11947/j.AGCS.2022.20220199.
	JIANG Bitao. The development and prospect of China's space earth observation technology[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(7):1153-1159.DOI: 10.11947/j.AGCS.2022.20220199.
[7]	李德仁. 从测绘学到地球空间信息智能服务科学[J]. 测绘学报, 2017, 46(10):1207-1212.DOI: 10.11947/j.AGCS.2017.20170263.
	LI Deren. From geomatics to geospatial intelligent service science[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(10):1207-1212.DOI: 10.11947/j.AGCS.2017.20170263.
[8]	赵文波, 李帅, 李博, 等. 新一代体系效能型对地观测体系发展战略研究[J]. 中国工程科学, 2021, 23(6):128-138.
	ZHAO Wenbo, LI Shuai, LI Bo, et al. Development strategy of the new-generation effectiveness-oriented earth-observation system[J]. Strategic Study of CAE, 2021, 23(6):128-138.
[9]	李双博, 张艳松, 李德仁, 等. 我国天基信息系统发展现状及未来展望[J]. 中国工程科学, 2020, 22(1):127-132.
	LI Shuangbo, ZHANG Yansong, LI Deren, et al. Development status and future prospect of space-based information systems in China[J]. Strategic Study of CAE, 2020, 22(1):127-132.
[10]	乔凯, 智喜洋, 王达伟, 等. 星上智能信息处理技术发展趋势分析与若干思考[J]. 航天返回与遥感, 2021, 42(1):21-27.
	QIAO Kai, ZHI Xiyang, WANG Dawei, et al. Analysis and some thoughts on the development trend of the on-board intelligent information processing technology[J]. Spacecraft Recovery & Remote Sensing, 2021, 42(1):21-27.
[11]	王密, 杨芳. 智能遥感卫星与遥感影像实时服务[J]. 测绘学报, 2019, 48(12):1586-1594.DOI: 10.11947/j.AGCS.2019.20190454.
	WANG Mi, YANG Fang. Intelligent remote sensing satellite and remote sensing image real-time service[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(12):1586-1594.DOI: 10.11947/j.AGCS.2019.20190454.
[12]	李德仁, 张洪云, 金文杰. 新基建时代地球空间信息学的使命[J]. 武汉大学学报（信息科学版）, 2022, 47(10):1515-1522.
	LI Deren, ZHANG Hongyun, JIN Wenjie. The mission of geo-spatial information science in new infrastructure era[J]. Geomatics and Information Science of Wuhan University, 2022, 47(10):1515-1522.
[13]	赵军锁, 吴凤鸽, 刘光明. 软件定义卫星技术发展与展望[J]. 卫星与网络, 2017(12):46-50.
	ZHAO Junsuo, WU Fengge, LIU Guangming. Development and prospect of software-defined satellite technology[J]. Satellite & Network, 2017(12):46-50.
[14]	徐帆江, 周鑫, 赵军锁, 等.软件定义卫星技术概念及发展[J].北京航空航天大学学报, 2023, 49(7):1543-1552.
	XU Fanjiang, ZHOU Xin, ZHAO Junsuo, et al. Rethinking of software-defined satellite technology [J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49(7):1543-1552.
[15]	李德仁, 王密, 杨芳. 新一代智能测绘遥感科学试验卫星珞珈三号01星[J]. 测绘学报, 2022, 51(6):789-796. DOI: 10.11947/j.AGCS.2022.20220184.
	LI Deren, WANG Mi, YANG Fang. A new generation of intelligent mapping and remote sensing scientific test satellite Luojia-3 01[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(6):789-796. DOI: 10.11947/j.AGCS.2022.20220184.
[16]	李德仁. 论空天地一体化对地观测网络[J]. 地球信息科学学报, 2012, 14(4):419-425.
	LI Deren. On space-air-ground integrated earth observation network[J]. Journal of Geo-Information Science, 2012, 14(4):419-425.
[17]	李德仁, 丁霖, 邵振峰. 面向实时应用的遥感服务技术[J]. 遥感学报, 2021, 25(1):15-24.
	LI Deren, DING Lin, SHAO Zhenfeng. Application-oriented real-time remote sensing service technology[J]. National Remote Sensing Bulletin, 2021, 25(1):15-24.
[18]	王时雨, 张盛兵, 黄小平, 等. 星载SAR成像与智能处理的单片多处理架构[J]. 西北工业大学学报, 2021, 39(3):510-520.
	WANG Shiyu, ZHANG Shengbing, HUANG Xiaoping, et al. Single-chip multi-processing architecture for spaceborne SAR imaging and intelligent processing[J]. Journal of Northwestern Polytechnical University, 2021, 39(3):510-520.
[19]	徐伟, 陈彦彤, 朴永杰, 等. 基于吉林一号遥感图像的星载目标快速识别系统[J]. 光学精密工程, 2017, 25(1):255.
	XU Wei, CHEN Yantong, PIAO Yongjie, et al. Target fast matching recognition of on-board system based on Jilin-1 satellite image[J]. Optics and Precision Engineering, 2017, 25(1):255.
[20]	周渊, 张洵颖, 智永锋, 等. 软件定义卫星有效载荷技术研究[J]. 西北工业大学学报, 2020, 38(S1):96-101.
	ZHOU Yuan, ZHANG Xunying, ZHI Yongfeng, et al. Research on software defined satellite payload technology[J]. Journal of Northwestern Polytechnical University, 2020, 38(S1):96-101.
[21]	吴启星. 软件定义卫星研究现状与技术发展展望[J]. 中国电子科学研究院学报, 2021, 16(4):333-337.
	WU Qixing. State of the art and development analysis of software defined satellites[J]. Journal of China Academy of Electronics and Information Technology, 2021, 16(4):333-337.
[22]	NELSON J A. Net centric radar technology & development using an open system architecture approach[J]. IEEE Aerospace and Electronic Systems Magazine, 2011, 26(5):34-37.
[23]	谢友柏.关于MBSE和MBD的思考[J].科技导报, 2019, 37(7):6-11.
	XIE Youbai.Thinking about MBSE and MBD[J]. Science & Technology Review, 2019, 37(7):6-11.
[24]	ANDERSON L, COLE B, YNTEMA R, et al. Enterprise modeling for CubeSats[C]//Proceedings of 2014 IEEE Aerospace Conference.Big Sky: IEEE, 2014.
[25]	COLE B, DELP C, DONAHUE K.Piloting model based engineering techniques for spacecraft concepts in early formulation[J].INCOSE International Symposium, 2010, 20(1):1351-1365.
[26]	MANDUTIANU S. Modeling pilot for early design space missions[C]//Proceedings of the 7th Annual Conference on Systems Engineering Research.Loughbo: IEEE, 2009: 20-23.
[27]	赵清, 林宝军, 张善从. JPEG2000遥感图像实时压缩系统[J]. 计算机仿真, 2007, 24(3):180-183.
	ZHAO Qing, LIN Baojun, ZHANG Shancong. JPEG2000 remote-sensing real-time encoding system[J]. Computer Simulation, 2007, 24(3):180-183.
[28]	王密, 项韶, 肖晶. 面向任务的高分辨率光学卫星遥感影像智能压缩方法[J]. 武汉大学学报（信息科学版）, 2022, 47(8):1213-1219.
	WANG Mi, XIANG Shao, XIAO Jing. Task-oriented intelligent compression method for high resolution optical satellite remote sensing image[J]. Geomatics and Information Science of Wuhan University, 2022, 47(8):1213-1219.
[29]	李劲东.卫星遥感技术[M]. 北京: 北京理工大学出版社, 2018: 499-501.
	LI Jindong. Satellite remote sensing technology[M]. Beijing: Beijing Insititute of Technology Press, 2018: 499-501.
[30]	白照广. 高分一号卫星的技术特点[J]. 中国航天, 2013, 424(8):5-9.
	BAI Zhaoguang. Technical characteristics of Gaofen-1 satellite[J]. Aerospace China, 2013, 424(8):5-9.
[31]	曹海翊, 张新伟, 赵晨光, 等. 高分七号卫星总体设计与技术创新[J]. 中国空间科学技术, 2020, 40(5):1-9.
	CAO Haiyi, ZHANG Xinwei, ZHAO Chenguang, et al. System design and key technolongies of the GF-7 satellite[J]. Chinese Space Science and Technology, 2020, 40(5):1-9.
[32]	SHEIKH H R, SABIR M F, BOVIK A C. A statistical evaluation of recent full reference image quality assessment algorithms[J]. IEEE Transactions on Image Processing, 2006, 15(11):3440-3451.
[33]	LUIGI DRAGOTTI P, POGGI G, RAGOZINI A R P. Compression of multispectral images by three-dimensional SPIHT algorithm[J]. IEEE Transactions on Geoscience and Remote Sensing, 2000, 38(1):416-428.
[34]	CHENG Gong, ZHOU Peicheng, HAN Junwei. Learning rotation-invariant convolutional neural networks for object detection in VHR optical remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(12):7405-7415.

处理器	GPU	FPGA	DSP	NPU
架构特点	基于单指令多数据流（SIMD）的架构	可基于特定算法设计专用硬件架构	适应并行的哈佛架构	可基于特定算法设计专用硬件架构
优缺点	生态好、算力高功耗高、散热难	高并行性、可重构开发效率低	稳定性高、接口全性能弱	能耗低、性能强架构固定
专用算力	高	动态算力	低	高
国产化程度	低	高	高	高
宇航级产品	无	有	有	待验证
本文选择的硬件组合（综合性能、功耗、国产化、宇航级成熟度）		√	√	√

个数	类型	卷积核数	尺寸	输出	运算量（OPS）	参数量
5	卷积	16/32/64/128/256	3×3/1	256×256	889 192 448	397 312
5	最大池化		2×2/2	128×128	889 192 448	397 312
1	卷积	512	3×3/1	8×8	150 994 944	1 179 648
1	最大池化		2×2/1	8×8	150 994 944	1 179 648
2	卷积	1024/512	3×3/1	8×8	805 306 368	6 291 456
2	卷积	256/45	1×1/1	8×8
1	卷积	128	1×1/1	8×8	4 202 496	32 768
1	上采样		×2	16×16
1	卷积	384	3×3/1	16×16	453 050 368	884 736
1	卷积	45	1×1/1	16×16	33 619 968	65 536

硬件平台	XC7VX690 T			XCVU9 P
硬件平台	单IP核/（Gb/s）	最大核数	总算力	单IP核/（Gb/s）	最大核数	总算力
云检测压缩	1.31	4	5.24 Gb/s（1.31 Gb/s×4）	1.31	6	7.86 Gb/s（1.31 Gb/s×6）
目标检测识别	0.22	6	1.32 Gb/s（0.22 Gb/s×6）	0.33	12	3.96 Gb/s（0.33 Gb/s×12）

处理需求	非模型定义方案			模型定义方案
处理需求	目标检测任务	图像压缩任务	算力利用率/（%）	目标检测任务	图像压缩任务	算力利用率/（%）
高分一号@1.28 Gb/s	1×690 T	1×690 T	39.02	5/6×VU9 P	1/6×VU9 P	55.53
高分七号@3.00 Gb/s	1×VU9 P	1×690 T	65.22	1/2×VU9 P＋1×690 T	1/2×VU9 P	82.99