Classification of Alpine Skiing Styles Using GNSS and Inertial Measurement Units

Abstract

In alpine skiing, four commonly used turning styles are snowplow, snowplow-steering, drifting and carving. They differ significantly in speed, directional control and difficulty to execute. While they are visually distinguishable, data-driven classification is underexplored. The aim of this work is to classify alpine skiing styles based on a global navigation satellite system (GNSS) and inertial measurement units (IMU). Data of 2000 turns of 20 advanced or expert skiers were collected with two IMU sensors on the upper cuff of each ski boot and a mobile phone with GNSS. After feature extraction and feature selection, turn style classification was applied separately for parallel (drifted or carved) and non-parallel (snowplow or snowplow-steering) turns. The most important features for style classification were identified via recursive feature elimination. Three different classification methods were then tested and compared: Decision trees, random forests and gradient boosted decision trees. Classification accuracies were lowest for the decision tree and similar for the random forests and gradient boosted classification trees, which both achieved accuracies of more than 93% in the parallel classification task and 88% in the non-parallel case. While the accuracy might be improved by considering slope and weather conditions, these first results suggest that IMU data can classify alpine skiing styles reasonably well.

Editorial: Health and Performance Assessment in Winter Sports

“…Recent advances in digitization and measurement technology have also influenced youth and elite winter sports. Current trends of health and performance assessment can be summarized as: (1) new technologies and approaches for motion analysis with a clear tendency toward wearables, large capture volumes and long-term assessments (Supej, 2010 ; Gilgien et al, 2013 , 2015 ; Fasel et al, 2017a , b ; Spörri et al, 2017 ; Fasel et al, 2018 ; Gløersen et al, 2018a , b ; Takeda et al, 2019 ; Neuwirth et al, 2020 ); (2) digital data collections, interdisciplinary/integrative measurement setups, advanced signal processing and state-of-the-art data analytics such as machine learning in connection with performance assessment and enhancement (Rindal et al, 2017 ; Jang et al, 2018 ; Losnegard et al, 2019 ; Ostrek et al, 2019 ; Skattebo et al, 2019 ; Heinrich et al, 2020 ; Solli et al, 2020 ); (3) mobile Health (mHealth) applications, supporting engineering solutions, and technology-based athlete screening in the context of injury/illness surveillance and prevention (Steenstrup et al, 2018 ; Schindelwig et al, 2019 ; Ellenberger et al, 2020a , b ; Franchi et al, 2020 ; Fröhlich et al, 2020 ; Hermann and Senner, 2020 ); and (4) comprehensive monitoring/test protocols, and the objectification of clinical criteria in the return-to-sport context (Jordan et al, 2015a , b ; Jordan et al, 2018 ; Csapo et al, 2019 ).…”

Human and Animal Motion Tracking Using Inertial Sensors

“…The second set of papers are research papers focusing on sports applications: soccer [4], ski [5], canoe [6] and racket playing [7].…”

“…The second sport-related research paper is focused on the use of inertial sensors to classify skiing practice and techniques [5]. In their paper, the authors present four skiing turn styles: snowplow, snowplowstreering, drifting and carving.…”

Comfortable and Convenient Turning Skill Assessment for Alpine Skiers Using IMU and Plantar Pressure Distribution Sensors

“…Furthermore, there have also been several previous ski-turn skill assessment studies. Motion analysis of ankle joints during ski turns has been performed by simultaneously using IMUs and plantar pressure distribution sensors [ 18 ]; skiing styles, such as skidding or carving turns, of alpine skiing have been evaluated using IMUs and the global navigation satellite system [ 19 ], and physical skills have been assessed from lower extremity muscle activity during skiing [ 20 ]. Although previous studies have provided detailed analyses and evaluations, these methods frequently use large and expensive technology, thus limiting their use by recreational skiers.…”