Abstract
Commercial devices for real-time sports performance measurement are available. The small size and low cost of micro-electro-mechanical systems (MEMS) inertial measurement units (IMUs) are exploited in these devices to make their body mountable and affordable for the general public. These devices are widely used in sports such as skiing, snowboarding and mountain biking to measure the air time (AT) of aerial athletic jumps in real time. AT is considered as an important performance variable. However, the performance of currently used real-time athletic jump AT determination algorithms using MEMS IMUs are unsatisfactory in terms of accuracy and reliability. In this study, a new algorithm for AT determination with a head-mounted MEMS IMU is proposed. The algorithm has a probabilistic approach using multiple attribute decision making. The concept of jump characteristic points and a novel method used in this algorithm, namely preceding and following acceleration difference, are introduced. The field experiment results demonstrate that the proposed algorithm comprehensively out-performs the commercial algorithms used in two state-of-the-art sport performance measurement devices. The proposed algorithm exhibits an average error of 0.033 s (4.8%), whereas the two commercial algorithms demonstrate average errors of 0.111 s (8.4%) and 0.538 s (39.7%).
Acknowledgements
This work was funded by the Natural Sciences and Engineering Research Council of Canada through an engage grant. This grant and the study performed under it received ethics approval for human participants by the UBC Okanagan Campus Behavioral Research Ethics Board (certificate H10-02174). The authors also wish to acknowledge Recon Instruments Ltd for their participation in this project.