Easily available information technology based on the data-driven models for social biomechanics

2023 Fiscal Year Final Research Report

• Project/Area Number: 20H04075
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 59020:Sports sciences-related
• Research Institution: Nagoya University
• Principal Investigator: Fujii Keisuke 名古屋大学, 情報学研究科, 准教授 (70747401)
• Co-Investigator(Kenkyū-buntansha): 石黒 祥生 東京大学, 大学院情報学環・学際情報学府, 准教授 (20769418)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 機械学習 / 集団運動 / マルチエージェント / 深層学習

Outline of Final Research Achievements

In this project, we conducted research using data-driven modeling with location data from team sports. The purpose of this research is to understand the functions and principles of team tactics and to develop a predictive model that incorporates tactical evaluation. In the first year, we developed a trajectory prediction model using imitation learning that takes into account soccer defensive tactics, and showed improvement in defensive indicators. From the following year onwards, we conducted multiple studies on movement prediction and player evaluation such as in badminton and soccer, which were accepted by several international conferences and journals. In particular, it is a new approach that estimates the player's action value function from data and evaluates decision making. These research results will contribute to advances in data analysis and modeling for team sports.

Free Research Field

スポーツ科学

Academic Significance and Societal Importance of the Research Achievements

本研究による集団スポーツのデータ駆動的モデリングは、チームの戦術的評価を取り入れた予測モデルの開発を実現し、スポーツ科学とデータ科学の横断的領域において重要な課題に取り組んだ。サッカーやバドミントンなどにおける戦術的予測モデルと選手評価は、コーチング戦略の最適化、選手パフォーマンスの向上、試合解析の精度を高めることに寄与することが期待される。また、意思決定プロセスの客観的評価手法の開発は、トレーニング方法の革新や戦術教育の向上にも貢献し、スポーツ界全体の競技レベルの向上を促進する社会的意義も大きいと考えられる。