Recent advances in the application of artificial intelligence and wearable devices in volleyball

Volleyball is characterized by high intensity, rapid pace, and multi-agent coordination, where performance depends not only on individual technical execution but also on collective behavior organization, offense-defense transition efficiency, and dynamic decision-making. Conventional volleyball performance analysis, which relies on expert judgment, manual statistics, and video replay, is constrained by labor-intensive analytical workflows and delayed feedback, limiting its ability to continuously and accurately capture complex behaviors during training and competition in real time. Moreover, the analytical outcomes cannot be translated promptly into targeted training interventions and evidence-based decision support. In recent years, advances in artificial intelligence and wearable technologies have emerged as promising solutions to these limitations, accelerating the transition of volleyball performance analysis from an experience-driven to a data-driven paradigm. This review systematically summarizes research progress in this field. It first examines the applications of machine learning and deep learning in motion recognition, event detection, behavior modeling, and outcome prediction. It then summarizes data acquisition foundations based on inertial measurement units, flexible sensors, and multimodal integration, and further synthesizes their applications in match and training analysis, athlete performance quantification and load assessment, training decision-making and tactical optimization, and match outcome prediction. Finally, the key challenges in current research are discussed, and future directions for the intelligent development of volleyball are outlined, providing a systematic reference for advancing intelligent volleyball analytics.