Wearable Technologies for Beach Volleyball Performance Improvement

This project aims to investigate to which extent wearable sensors can be used to analyse and improve the performance of the beach volleyball players. Nowadays the usage of wearable devices in sport has became widespread thanks to commercial activity and fitness tracker. While most of them are oriented on the most common study of activities like running, cycling, etc. Several studies investigate also the usage of wearable sensors for performance analysis in sports like baseball, swimming, golf, but there is no previous work that has investigated the application of these sensors in order to analyse and improve beach volleyball athletes’ performance.

Beach volleyball presents several challenges in the performance analysis using wearable sensors: the set of possible gestures is wide and each gesture can be performed differently by each player. Each gesture is also performed very quickly during the game and the execution of each one of them could be influenced by the interactions between the team mates. Finally, the irregularity of the ground requires robustness in the analysis, because the sensors data can be influenced when the players move on the sand.

To tackle these challenges, we will develop this project in two steps:

• In the first phase, we aim to improve the quality of the execution of the gesture of the player during the traning;
• In the second phase, the performance will be analysed and evaluated during a game, instrumenting all the four players on the court.

Single technique improvement

During this phase, we aim to analyse and improve the performance of the players by using 12 wearable sensors. Each one of these sensors is a custom made wireless platform integrating a 9-axis inertial unit called BlueSense. These sensors and pattern recognition algorithm will be used during this project in order to classify each technique and to provide a feedback for the players according with this classification.

Game Performance Improvement

In the second phase of the project, the goal will be to analyze the performance of the players during a game. By instrumenting 4 players with 3 BlueSenses each, we aim to classify each gesture performed during the game. We want to analyse the outcome of the gestures in term of effort required by the other players to respond that action. In order to address this, several machine learning techniques will be explored.

Publications

• Cuspinera, L. Ponce, et al. "Beach volleyball serve type recognition." Proceedings of the 2016 ACM International Symposium on Wearable Computers. ACM, 2016.
• Roggen, D., Cuspinera, L. P., Pombo, G., Ali, F., & Nguyen-Dinh, L. V. (2015, February). Limited-memory warping LCSS for real-time low-power pattern recognition in wireless nodes. In European Conference on Wireless Sensor Networks (pp. 151-167). Springer International Publishing.

People

Mathias Ciliberto, Dr Hristijan Gjoreski, project supervised by Dr.Daniel Roggen, Dr. Luis Ponce Cuspinera.