Elucidating Joint Motions to Improve Canoe Paddling: Application of Machine Learning to an Instructional System

2021 Fiscal Year Final Research Report

• Project/Area Number: 20K19573
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 59020:Sports sciences-related
• Research Institution: Hokkaido University of Science
• Principal Investigator: Nomura Tomohiro 北海道科学大学, 保健医療学部, 准教授 (90593492)
• Project Period (FY): 2020-04-01 – 2022-03-31
• Keywords: カヤック / パドリング / カヌー / 教示システム

Outline of Final Research Achievements

We measured and analyzed the paddling behavior of a former Japanese national canoe sprint kayaker using a three-dimensional motion analyzer. From the results, we defined new phases and indices that were considered to have a significant effect on performance. To confirm this, we measured and analyzed data from five experienced kayakers, and found that there was a strong correlation between the competition time and the defined indices.
We developed a paddling motion teaching system using this index. One canoe ocean racer practiced using the system for three days, and the results were very good: his 7-km time was reduced by 25 minutes.

Free Research Field

スポーツ科学

Academic Significance and Societal Importance of the Research Achievements

カヤックのパフォーマンス評価について，いくつか指標が作られているが，実際の競技タイムと適合しない場合が多々存在する．本研究では，パドリング動作に着目して，新たな評価指標を提案し，その有用性を確認した．また，この指標を組み込んだ教示システムを開発し，短期間で競技タイムを大きく縮めることに成功した．今後は，この指標及びシステムを併用して練習していくことで，課題の可視化によるモチベーション向上や，効率的なパフォーマンスの向上に役立つものと考えられる．