arXiv cs.AI· Pedro Antunes, Artur Podobas·· 3 天前AI 评分15
面向太空应用的 FPGA 神经网络加速器综述
FPGA-Based Neural Network Accelerators for Space Applications: A Survey
AI 导读
一篇 48 页综述分析了面向太空任务的 FPGA 神经网络加速器研究现状，指出 FPGA 因灵活性、成本效益和抗辐射潜力受到关注，可用于自主运行、传感器数据分析和数据压缩等星上任务。该文梳理了现有文献的趋势与空白并提出未来研究方向，已于 2026 年 10 月 7 日在线发表于 ACM Computing Surveys。
正文 · 原文
Abstract:Space missions are becoming increasingly ambitious, necessitating high-performance onboard spacecraft computing systems. In response, field-programmable gate arrays (FPGAs) have garnered significant interest due to their flexibility, cost-effectiveness, and radiation tolerance potential. Concurrently, neural networks (NNs) are being recognized for their capability to execute space mission tasks such as autonomous operations, sensor data analysis, and data compression. This survey serves as a valuable resource for researchers aiming to implement FPGA-based NN accelerators in space applications. By analyzing existing literature, identifying trends and gaps, and proposing future research directions, this work highlights the potential of these accelerators to enhance onboard computing systems.
| Comments: | 48 pages, 4 figures, 4 tables. Author's accepted manuscript, published online in ACM Computing Surveys on 7 October 2026. This version replaces the April 2025 preprint |
| Subjects: | Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI) |
| ACM classes: | B.7.1; C.3; I.2.6; J.2 |
| Cite as: | arXiv:2504.16173 [cs.AR] |
| (or arXiv:2504.16173v4 [cs.AR] for this version) | |
| https://doi.org/10.48550/arXiv.2504.16173 arXiv-issued DOI via DataCite |
|
| Related DOI: | https://doi.org/10.1145/3857797
DOI(s) linking to related resources |
Submission history
From: Pedro Antunes [view email]
[v1]
Tue, 22 Apr 2025 18:02:35 UTC (7,368 KB)
[v2]
Thu, 24 Apr 2025 12:04:11 UTC (7,368 KB)
[v3]
Fri, 12 Jun 2026 16:08:09 UTC (637 KB)
[v4]
Mon, 12 Oct 2026 14:11:14 UTC (712 KB)
来源：arXiv cs.AI · arxiv.org