Genomförbarhet av IMU-baserad mätning av träningsbelastning i trampett inom truppgymnastik

Wearable sensor technology is transforming sports analysis by enabling detailed, real-time motion tracking outside of laboratory settings. In gymnastics disciplines like TeamGym, where athletes perform rapid and complex movements, accurately monitoring training load remains a significant challenge. Effective training load monitoring is essential for optimizing performance and reducing injury risk, but traditional motion capture tools are often not feasible for in-field use in gymnastics environment. The goal of this thesis is to systematically evaluate whether IMU can provide reliable and meaningful training load metrics for TeamGym trampette jumps. Key steps included identifying relevant features of external training load, implementing IMU-based methods for measuring them, validating their accuracy, reliability and sensitivity, and developing a mobile application to support coaches and athletes. Data was collected both through test data protocol and real world data protocol, and signal processing methods and algorithms were used to extract horizontal velocity, peak jerk, net impulse, jump height, total energy and jump count. The results show that peak jerk, net impulse, jump height, total energy at jump apex and jump count, are promising indicators of external training load. The IMU-based methods where validated against video-based methods to further strengthened their credibility. A prototype app was successfully developed to deliver jump count feedback and demonstrated potential for integrating additional features. These findings highlight the feasibility of using IMUs for performance monitoring in TeamGym and lay the groundwork for technology-assisted, data-driven coaching. With further development, such systems may also help reduce injury risk by detecting excessive load patterns and supporting more informed training decisions.