INTRODUCTION: Some descriptive biomechanical parameters have been identified to be related to ski jumping performance (e.g. lift force). However, measuring these parameters is a non trivial task which requires complex devices, well trained staff and time consuming post-processing. Therefore, these parameters are almost never considered in daily training, but rather limited to research projects. The first aim of this study was to propose a system to capture the athlete movement during the entire jump that is compatible with daily training requirements. The second aim was to assess the relevance of angular, coordination and dynamic parameters automatically calculated by the proposed system. METHOD: The system was composed of eight small inertial measurement units (IMUs), including three-dimensional (3D) accelerometer and gyroscope (Physilog®, CH). The IMUs were attached to skis, shanks, thighs, sacrum and torso. 32 male athletes from the Swiss ski team, from juniors to world-class athletes, were asked to perform up to three jumps while wearing the system. Before each jump, a functional calibration was realized to align the IMUs to body frames. The jump phases were identified according to Aminian et al. (2009). Skis and Segments 3D angular movements were computed during the entire jump (Aminian, 2010). In addition, inter-segmental coordination (continuous relative phase (CRP)), 3D center of mass dynamics and vertical ground reaction force (GRF) were calculated during take-off (TO). Finally, aerodynamic forces were estimated during stable flight (SF). To compare jumps, features were extracted from the angular, coordination and dynamic curves. Regression analysis was used to assess the relations between extracted parameters and distance. RESULTS AND DISCUSSION: This first application of the system confirmed its compatibility with daily training requirements. The angular, coordination and dynamic features were automatically extracted for 86 jumps in total. This analysis pointed out strong interactions between parameters and between parameters and distance. For shake of consistency, only four observations are discussed below. First, the maximum GRF and mean somersault angular velocity during TO showed a medium correlation with distance. Second, a large hip angle at the end of TO and a static shank segment at the beginning of TO (CRP) expressed a medium correlation with distance, indicating that the best athletes tend to fix their shank and open their upper body. Third, the mean thigh angular velocity during TO reported a high correlation with distance, GRF and somersault angular velocity. This highlights the importance of thighs movement. Fourth, several kinematic (e.g. ankle flexion angle, skihorizontal angle) and aerodynamic parameters of early-flight and SF were significantly correlated with distance. Finally, based on the multiple regression, the most relevant metrics could explained 76% of the distance variance. CONCLUSION: This system allowed capturing the movement during the entire jump. Moreover, it automatically extracted a large number of angular, coordination and dynamic parameters, out of which several were directly correlated with distance. In conclusion, this study presented a simple method that can be used to study ski jumping or to provide a rapid feedback to athletes and coaches for a more efficient training.
© Copyright 2012 Science and Skiing V. 5th International Congress on Science and Skiing, Dec. 14 - 19, 2010, St. Christoph am Arlberg. Veröffentlicht von Meyer & Meyer Sport (UK) Ltd.. Alle Rechte vorbehalten.