EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI-Enhanced Sports Training in Physical Education: Global Research Trends, Pedagogical Methods, and Ethical Frameworks (2002–2025)

Mengshi Li, Binbin Zhang, Gaopeng Li, Weiming Tao, Mohd Taib Harun, Jian Li, Jinglang Fu

Physical Education Theory and Methodology · 30.03.2026 · 26 (2) · 251-262

DOI: 10.17309/tmfv.2026.2.03 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physical Education Theory and Methodology, online erschienen am 30.03.2026. Verfasst von Mengshi Li und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Artificial intelligence (AI) is rapidly entering physical education and sports training, providing new avenues for motion observation, learning diagnosis, and feedback. However, the pedagogical orientation and classroom feasibility of related research remain obscured by narratives focused primarily on technological performance.

Objectives

This study reviews and characterizes the knowledge base, thematic structure, and evolutionary trajectory of global AI-enabled physical education research from 2002 to 2025. It highlights implications for physical education decision-making and ethical governance, emphasizing the principle of “assisting teachers rather than replacing them.” Materials and Methods. A Web of Science Core Collection search conducted in April 2025 yielded 729 peer-reviewed English-language journal articles. Co-citation networks, keyword co-occurrence analysis, and BERT-based semantic clustering were used to identify core literature, thematic clusters, and stages of development. These results were integrated with an interpretive synthesis from a physical education perspective.

Results

Research has evolved from early motion quantification and data-driven analysis to deep learning-based motion recognition and generative AI-supported instructional feedback, demonstrating a shift toward “instructional-ethical governance.” However, the evidence primarily focuses on universities and elite training programs. Empirical evidence in K–12 classrooms remains limited, and replicable implementation processes are lacking, creating a “laboratory–classroom” transition gap. Furthermore, issues such as curriculum articulation, learning stages, feedback timing, classroom risk management, and privacy and fairness remain insufficiently addressed.

Conclusions

In physical education, AI is best positioned as a teacher-led decision-support tool that enhances observation, formative assessment, and diagnostic feedback. Teachers should retain responsibility for key instructional judgments, prioritizing athletic development while ensuring safety and fairness. This study recommends using teaching effectiveness as the primary evaluation benchmark to promote explainable, controllable, and compliant classroom implementation pathways.

Zitieren

APA

Li, M., Zhang, B., Li, G., Tao, W., Harun, M. T., Li, J., & Fu, J. (2026). AI-Enhanced Sports Training in Physical Education: Global Research Trends, Pedagogical Methods, and Ethical Frameworks (2002–2025). Physical Education Theory and Methodology, 26(2), 251-262. https://doi.org/10.17309/tmfv.2026.2.03

Vancouver

Li M, Zhang B, Li G, Tao W, Harun MT, Li J, et al. AI-Enhanced Sports Training in Physical Education: Global Research Trends, Pedagogical Methods, and Ethical Frameworks (2002–2025). Physical Education Theory and Methodology. 2026;26(2):251-262. doi:10.17309/tmfv.2026.2.03

BibTeX
@article{Li2026,
  author = {Li, Mengshi and Zhang, Binbin and Li, Gaopeng and Tao, Weiming and Harun, Mohd Taib and Li, Jian and Fu, Jinglang},
  title = {AI-Enhanced Sports Training in Physical Education: Global Research Trends, Pedagogical Methods, and Ethical Frameworks (2002–2025)},
  journal = {Physical Education Theory and Methodology},
  year = {2026},
  volume = {26},
  number = {2},
  pages = {251--262},
  doi = {10.17309/tmfv.2026.2.03}
}
Kostenlos starten →