研究人员开发了Star-Fusion，这是一种新颖的多模态Transformer架构，用于航天器导航中的精确天体定向确定。该方法将问题重新构建为离散拓扑分类任务，利用球形K-Means聚类来处理天体球面非欧几里得拓扑的复杂性。该架构集成了SwinV2-Tiny Transformer进行特征提取，卷积热图分支进行空间定位，以及MLP进行几何锚定。实验结果表明，Star-Fusion的Top-1准确率达到93.4%，推理延迟为18.4毫秒，适合实时在轨卫星应用。 AI
影响 这种新架构可以提高自主航天器导航系统的准确性和效率。
排序理由 这是一篇描述针对特定技术问题的模型新架构的研究论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
Reliable celestial attitude determination is a critical requirement for autonomous spacecraft navigation, yet traditional "Lost-in-Space" (LIS) algorithms often suffer from high computational overhead and sensitivity to sensor-induced noise. While deep learning has emerged as a p…
arXiv:2604.26582v1 Announce Type: new Abstract: Reliable celestial attitude determination is a critical requirement for autonomous spacecraft navigation, yet traditional "Lost-in-Space" (LIS) algorithms often suffer from high computational overhead and sensitivity to sensor-induc…
Reliable celestial attitude determination is a critical requirement for autonomous spacecraft navigation, yet traditional "Lost-in-Space" (LIS) algorithms often suffer from high computational overhead and sensitivity to sensor-induced noise. While deep learning has emerged as a p…