Real-Time Motion Prediction of Small High-Speed Vessels Using Data Assimilation and Learning Theory

2023 Fiscal Year Final Research Report

• Project/Area Number: 20H02381
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 24020:Marine engineering-related
• Research Institution: 防衛大学校(総合教育学群、人文社会科学群、応用科学群、電気情報学群及びシステム工学群)
• Principal Investigator: Terada Daisuke 防衛大学校(総合教育学群、人文社会科学群、応用科学群、電気情報学群及びシステム工学群), システム工学群, 教授 (80435453)
• Co-Investigator(Kenkyū-buntansha): 平川 嘉昭 横浜国立大学, 大学院工学研究院, 准教授 (00345480) ; 片山 徹 大阪公立大学, 大学院工学研究科, 教授 (20305650)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: 海難防止 / 海上安全

Outline of Final Research Achievements

The purpose of this study was to develop a real-time prediction method for ship motion, which is an elemental technology for the development of a safety operation support system for small high-speed vessels. The motion was predicted by estimating the external forces such as waves acting on the hull by successive data assimilation, modeling these external forces based on machine learning theories such as recurrent neural networks and radial basis neural networks, and giving the time series of external forces predicted by this model as the driving force for the simulation model used in successive data assimilation. This was done by giving the time series of external forces predicted by this model as the driving force of the simulation model. It was found that the prediction performance depends on the phase information of the estimated external force. In addition, the development of a new model experiment system for small high-speed vessels was realized.

Free Research Field

船舶海洋工学

Academic Significance and Societal Importance of the Research Achievements

逐次データ同化と学習理論を融合することにより、不規則な運動の予測が物理モデルに基づいて実現できることを示した点は学術的な意義が大きい。また、予測性能を決定づける要因が特定できたことから、今後の研究・開発の進展が見込める。本研究課題の成果を応用することにより、操船者の操船行為に起因する小型高速船の海難は未然に防止できることから、社会的な意義も大きい。