Reliable jump detection for snow sports with low-cost MEMS inertial sensors

Abstract

Body-mounted devices, incorporating low-cost micro-electromechanical systems (MEMS) Inertial Measurement Units (IMUs), for real-time sports performance feedback are commercially available. In sports such as skiing, snowboarding, and mountain biking, aerial jumps can be detected with these devices and performance variables including air time and jump drop can be calculated real-time. However, the performance of currently used real-time athletic jump detection algorithms using MEMS IMUs is unsatisfactory in terms of accuracy, power efficiency, and reliability. In this paper, a novel algorithm for jump detection with a head-mounted MEMS IMU is proposed. Two novel methods used in this algorithm, namely Windowed Mean Canceled Multiplication and Preceding and Following Acceleration Difference, are introduced. Field experiments are conducted and the results of the proposed algorithm are compared with those of algorithms used in two state-of-the-art sport performance measurement devices. Results demonstrate that the proposed jump detection algorithm comprehensively outperforms these commercial algorithms.

Keywords:

• Jump detection
• inertial sensors
• MEMS

Acknowledgements

This work was funded by the Natural Sciences and Engineering Research Council of Canada through an Engage grant. This grant and the work 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.