Journal DOI 10.68956/journal.human-performance-futures

Creative Commons source reading · Machine learning for performance

Machine learning–assisted triboelectric nanogenerator technology for intelligent sports

Ji M; Wang Z; Wu J; Huang L; Zheng M; Cheng G; Cai H; Luo J; Zhou H; Wang ZL

Originally published in Sci Adv · 2025-10-01

From the article

Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.

Abstract

AbstractThe rapid development of internet of things, big data, and artificial intelligence is propelling sports science into a data-driven era, demanding real-time, multidimensional athletic performance monitoring. Triboelectric nanogenerators (TENGs) have demonstrated exceptional potential in intelligent sports. However, the complexity and volume of TENG-generated data pose challenges for manual analysis. Machine learning (ML), with strengths in pattern recognition and adaptive processing, provides a powerful solution to enhance TENG-based sensing signal interpretation. This review systematically explores the integration of ML and TENG technology for intelligent sports. First, the fundamental theory and basic knowledge of TENGs are introduced, highlighting their versatility in sports sensing systems. Subsequently, a comprehensive…

DISCUSSION AND OUTLOOK

Intelligent sports facilities and wearable equipment are increasingly adopted in the sports industry, driving the trend toward digitalization and intelligent systems (179–181). Maintenance-free, sustainable, and highly sensitive sensors are essential for acquiring and analyzing large-scale, real-time motion data. In this review, we have summarized recent advancements in TENG technology for constructing self-powered intelligent sports systems, with a particular emphasis on its integration with ML. The integration of TENGs with ML presents a promising approach for enhancing intelligent sports applications, especially in the domains of real-time motion sensing, adaptive feedback, and data-driven decision-making. By leveraging the flexible, self-powered, and high-sensitivity features of TENGs alongside the data processing and pattern…

Editorial reading guide

  • What data created and tested the model, and was evaluation independent?
  • What is the cost of a false positive, false negative or out-of-distribution decision?