Automated exercise classification for wearable-sensor based weight training performance monitoring

点击翻译按钮获取中文摘要

Push Inc. is a Canadian startup based in Toronto, Ontario, focused on the development of wearable fitness devices for athletic weight training. Push has developed a wearable fitness device and accompanying mobile app that allows athletes to track their weight lifting exercises and compute performance metrics such as velocity, power, force and mechanical work, which they can use to monitor their training performance and progress. In the current implementation of the system, users must manually enter the exercise type and load into the app, to allow the system to know which exercise is currently being performed. If the system could detect the exercise being performed automatically, this would eliminate the need for the user to specify them manually, increasing usability and convenience, and reducing manual input errors. The aim of the proposed research is to develop the data processing and classification algorithms to enable the system to automatically detect the type of exercise being performed from the measured movement.