Towards Risk Indication In Mountain Biking Using Smart Wearables
Stefan Langer, Dennis Dietz, Andreas Butz
Abstract
Mountain biking as a recreational sport is currently thriving. During the ongoing COVID-19 pandemic, even more people started compensating for a lack of activity through individual outdoor sports, such as cycling. However, when executed beyond paved forest roads, mountain biking is a sport with subjective and objective risks, in which crashes often can not be entirely avoided and athletes may get injured. In this late-breaking work, we showcase a concept for a crash risk indication application for sports smartwatches. First, we review a wide range of related work, which formed the basis for our crash risk indication metric. We discuss options for the sensor-based detection of internal and external risk factors and propose a way to aggregate them, which will allow dynamic and potentially automatic fine-tuning by observing or obtaining feedback from the athlete. In addition, we present a concept for a smartwatch application that will provide constant feedback and an unobtrusive signal to the athlete when an unusually high risk is detected. Finally, we give an outlook on the necessary steps to implement our approach as a smartwatch app.