Data-driven approach for understanding chaotic phenomena in spacecraft orbital mechanics

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K18781
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 24:Aerospace engineering, marine and maritime engineering, and related fields
• Research Institution: Kyushu University
• Principal Investigator: Bando Mai 九州大学, 工学研究院, 教授 (40512041)
• Project Period (FY): 2021-07-09 – 2024-03-31
• Keywords: 多体問題 / カオス / データ駆動 / バリスティックキャプチャ / 宇宙ミッション

Outline of Final Research Achievements

In this study, we combine data-driven methods such as dynamic mode decomposition and delay embedding to generate a model that recovers and predicts the ballistic capture trajectory of a spacecraft multi-body dynamical system. The results show that for a finite number of time steps, it is possible to construct a model that can predict chaotic transition phenomena with sufficient approximation accuracy by using a sufficient number of data. Furthermore, based on the findings from the data-driven analysis and the dynamical system analysis, we investigated the dynamical system indices that characterize the ballistic capture trajectory, and examined the relationship between the time until the ballistic capture occurs and FTLE.

Free Research Field

アストロダイナミクス

Academic Significance and Societal Importance of the Research Achievements

本研究により得られた知見は，宇宙工学，天体力学の惑星形成論と異なる研究領域にまたがる多体力学系のカオス的現象について，統一的な解釈を可能とする.そして，現象のメカニズムをデータ駆動型アプローチでモデル化することにより，そのメカニズムの積極的な抽出が可能となり，宇宙ミッション策定の際の軌道設計への応用が期待される.また，データ駆動型サイエンスの非線形力学系のカオス現象への予測精度の限界などを明らかにすることは，他の学問分野においても一般化でき重要な意味を持つ.