Performance in rowing is reliant on the precise and consistent execution of strokes.Perfection of the stroke takes years of focused effort, with particular focus on the timing of thelegs, trunk and arms. Training is required not only on the water but on stationary training ergometers. Currently, instrumentation of the rowing stroke timing elements is limited to ratingand coaching feedback. There is still no componitized feedback system for the three mainelements of the stroke: legs, body, arms. This project`s aim is to create a visualization tool for rowers that breaks down the three movements in the stroke to provide coachable feedback. The tool has three microelectromechanical system (MEMS) sensors, each of which is positioned corresponding to the targeted movement. The first sensor is located on the seat of the erg, the second on the handle, and the last sensor on the rower`s back. A display of the rower`s timing element data allows them to learn and correct their strokes in real time. The system is still in development so it has a few limitations. A bad connection from the tape inhibited data transfer so heavy interpolation was used. Underfitting is likely. Possible future routes for this project include using machine learning to determine stroke quality, using machine learning to detect the phases of the stroke, engineering a wearable holder for the sensors to improve data transfer, and developing better ways to visualize the stroke in real time.
© Copyright 2020 Published by Faculty of Worcester Polytechnic Institute. All rights reserved.